Forecast Of The Capitalisation Yield

Published Date: 02 Nov 2017

Submitted By: Michal Krajčovič, username 090780mk

Submitted To: Dr. Wang Hui

Dated : 19.3.2013

The proposal should be 2500 words: +/- 10%.

word count ??? (literature list, annexes/figures and footnotes not included in word count)

Title

Real estate office market in Prague – Analysis and forecast of the capitalisation yield and rental levels

Background and research problem definition

Real estate has always been an essential part of life and economy, far before shares or bonds or securitisation was introduced. It can be used as consumer good and investment good, and additionally it can use these purposes at the same time (DiPasquale and Wheaton, 1996). This is valid considering a long term capital appreciation of real estate, and especially land [1] .

In most of developed and developing countries real estate is one of the most significant parts of peoples’ property, or 1/3th of the value of all capital assets in the world (Michael & Lizieri & Macgregor, 1998). Due to the fact that it is (i) significantly interlinked with nearly all parts of the economy and (i) society, and it is (i) fixed and unmovable it is ixtremely interesting for polititians on on the national and local (property taxes and how to structure them, regulation of urban development, infrastructure, agricultural sector,

Due to these features it has been considered as a specific sector with its unique economic characteristics. Real estate is closely linked to almost all sectors of the economy having a big influence on its micro structure and macro evolvement. The real estate is not only a product, good or investment asset but it is also a powerful tool in politician’s hands with a vast social implication.

The convenient characteristic of a property is its materiality (it is real). Unlike many of other (financial) capital assets you can literally touch the real estate. The realness (materiality) and durability have plenty of advantages. It is generally easy to prove its existence but above all, there exists a psychological value which can take many forms as well (Hoesli, 1993) [2] .

The term real estate can be defined from many perspectives. Generally we can define as land and its appurtenances that are built on it (Geltner and Miller 2007). But According to the same authors we can differentiate between the physical, the legal, and the financial economic perspective. (i) The physical "Bricks-and-Mortar Concept" defines real estate as a three dimensional structure of walls, ceilings, and floors. However, as Schulte et al. (2000) state, this definition does not consider aspects of the underlying parcel. (ii) From the legal point of view, real estate can only be regarded as a building, stationary and fixed at a certain location, in combination with a parcel and the assigned rights. The judicial definitions vary throughout jurisdictions. (eg is building legally separable and autonomous from the land or not, what is the land ownership and usage title [3] ). (iii) The economic definition according to Schäfers (1997) focuses on real estate (i) as a factor of production, and (ii) as real or capital investment, with a potential future cash-flow income.

Real estate properties are divided into several categories; based on their primary use

and the attractiveness or quality of the property itself. Very basic division is according

to their predominant type of use to 1) residential properties and 2) commercial

properties.

Residential property is defined as a type of land use where the predominant use is housing. Commercial property (also called investment or income property) are defined as buildings or land intended to generate a profit, either from capital gain or rental income (Investopedia 2013).

Commercial properties can be categorised as

Office properties – in larger cities are formed in clusters [4] , which are - downtown offices and Central business districts (CBD [5] ) offices. Office real estate is the biggest commercial property sector.

Retail (shopping) properties - Retail properties can be further divided into following categories (Geltner 2007):

High street

Convenience centres

Neighbourhood shopping centres

Community shopping centres

Regional shopping centres/ malls

Superregional shopping centers/ malls

Specialty centers

Lifestyle centers

Off-price outlets and discount centers/ malls

Highway commercial centers

hotel&leisure

logistics

industrial - Industrial properties include warehouses and manufacturing plants which are usually large one storey buildings. Due to their vast land requirement industrial properties are usually built outside of the city centres and close to main roads

mix use – mostly high street downtown including retail in the ground and either office or residential in the upper floors. In a strict wording most office development are actually mixed use (most new office buildings have some retail element the ground floor). In practice mixed use label is used when significant amount is generated at least by two property types (residential flats + retail in the ground floors; office + shopping mall; office + hotel+congress halls, etc.)

Office real Estate investments

In comparison with other assets real estate investment has been historically seen as a relatively safe long-term type of investment. At the same time it is still able to generate higher returns than long term government bonds This often presented perspective is partially skewed because securitised assets are compared to non-securitised assets. R/E mostly non-securitised thus highly illiquid and also without any possibilities how to measure net performance. Net performance on a stock or bond is easily computable on Bloomberg, in case of non-securitised R/E performance is often actually a gross performance without transaction costs, fees, operational costs etc.

Institutional and private investors perceive real estate assets as relatively safe type of an investment. Therefore it is often used to diversify an investment portfolio. Modern portfolio theory suggests investing into combination of assets that do not correlate perfectly (Modern Portfolio Theory developed by Markowitz). Such a combination has the ability to reduce risk while maintaining the same level of returns. [1] [2]

Commercial real estate investors are usually institutional investors or high net worth individuals (HNWI).

Investments into commercial R/E can be done directly or via other entities like SPVs [6] , REITs, pension funds, investment funds, etc. (done by institutional investors and retail individuals). Via these institutional intermediaries retail investors are investing in commercial real estate.

Real estate value is derived from current local market condition, as it is given by its

definition of an asset permanently connected to ground. Determinants such as local

supply and demand are major factors influencing the value of real estate. Constraints

such as zoning, planning or urban policy can attract or drive investors away.

The main differences (difficulties) of non-securitised R/E compared to traditional investment assets, like bonds and stocks, are the following:

Immobility - Real estate is immobile (thus Immobilien in German language and other translations of the word real estate). Real estate is irremovably connected with its land.

Heterogeneity – any two projects may have differing location, structure, age, architecture, (when considered within an SPV as is usual, we can add differing lease contracts, bank loan, supplier contracts) etc. There do not exist two same (absolute interchangeable) buildings. There could be two buildings having same size, same construction and same age neighbouring each other, but there will always be some differences (at least they cannot stand on the same plot). This heterogeneity is similar for example to private equity investments.

High transaction costs – commercial are usually larger constructions and represent larger investments (tens to hundreds of million EUR). During the M&A transaction a presence of third party advisors on both sides is common for the Due diligence (DD) and negotiations with the counterparty i.e. agents, lawyers, technical advisors, financial and tax advisors, etc.). The transaction costs can reach several per cent (1% - 4% in total) on each side (compared to several bips in case of securitised trades).

Long transaction time – DDs, negotiations, communication with authorities cause RE transactions to take months to even years to finish. Even then as closing is finished still certain obligations (such as reps and warranties) remain especially from the seller. Compared to

Small number of transactions – this drawback is closely related to the high transaction costs and long time to settle. The relatively small number of transactions for an asset type cause a problem of the information function of the market and by that slower (or even hinder) discovering the optimal price. In periods of strong buyers’ market (very small number of potential buyers) sometimes none or only a few relevant transactions occur in a given year. Compared to liquid markets where market makers offer a price all time (when the market is open). Price can then be created by a single investor (e.g. local private buyer Czech Property Investments in 2011)

Rigidity of supply (sticky prices) – sellers seem to be very unwilling to trade for prices cheaper than the amount they have paid for the property because it leads to significantly inelastic prices in the downside direction due to the durability nature that allows postponing the transaction to the ―better time (Case & Glaeser & Parker, 2000). Short sales and other tools of derivative markets which would allow making profit also in case of price decrease are generally not used in the direct real estate market.

Imperfect information – this issue is fundamentally connected with markets trading non standardised goods (eg. M&A). Lack of information on the buyer’s side can lead to the buyer being exploited or to adverse selection. These problems are mitigated by the involvement of a third party advisors. Although the absolute service fee amount is high, it is quite low relative to the value of traded property (1-4%) and generally can safe multiple amounts. It is often mentioned that there is never enough due diligence.

Long market adjustment mechanisms - it can take two to five years from the planning to the completion and letting of a real estate project. This can in some cases be a period of one economic cycle. IN case of a unexpected shift in the demand curve (positive or negative), a time lag can be observed until the demanded new office premises space come to the market [7] .

Commercial Office space and office market

Technical quality of the building:

The Urban Land Institute and The Building Owners and Managers Association (BOMA) , both noted authorities on commercial land uses, classifies office space into three categories: Class A, Class B, and Class C. The Urban Land Institute defines the classes as follows.

Class A space can be characterized as newer buildings that have excellent location and access, attract high quality tenants, and are managed professionally. Building materials are high quality and rents are competitive with other new buildings.

Class B buildings are typically smaller, and older but have good locations, management, and construction, and tenant standards are high. Buildings should not be out-dated and show very little deterioration. If the buildings are newer then they are typically smaller and not in a prime location.

Class C buildings are typically 15 to 25 years old but are maintaining steady occupancy. Their condition is typically fair but not considered good.

Overall quality of the property:

Prime property - Refers to property which is the best in terms of rentals, location, etc. I.e. is a newly completed A class building, with has creditworthy tenants and is on prime location [8] .

In transaction and rental statistics prime properties are usually used for measurement. Data for non-prime properties is (especially in small markets like Prague) non standardised and not consistent [9] . In my model I will also data referring to prime properties Prime properties (prime headline rent, initial yield on prime properties)

Short Historical background of the Czech economy and the Prague office market

Before 1989 - Centrally planned economy with a limited number of office jobs

After 1989 - Massive inflow of foreign direct investments (FDIs) and portfolio investments. Large privatization and changes in the legal framework made cross border investment possible. Large number of western companies set up branches. As of the early 2000s’a consolidation of the market is visible where western companies which were not profitable in CR cancel their branches in Prague due to low market share and limited growth potential (this continues till today).

2004 - Czech Republic (plus other CEE countries) entered European Union in May.

2004 - 2007 Oversupply of liquidity. Large supply of new offices, low interest rates, low yields, rising or stable effective rents

2008 – after the Lehman brothers CH11 filling the credit crunch and recession spread to central Europe as well, Investors had no interest in Prague market, run to quality/safe-heavens was in place. Stopped new construction of offices

2009 - economical recession. No new construction of offices, decreasing construction costs.

2010 - 2011 Slight Recovery. Limited new construction of offices, decreasing construction costs, start of the decrease of effective rents in EUR.

2012 - 2013 Double dip (W) and continuing stagnation / recession. Limited new construction of offices, continuing decrease of effective rents in EUR.

Motivation and reasoning behind the topic

As in every market all market participants are interested in an answer to the question what will be the prices in the future. This is the determinant of their success od failure as market participants. Thus, having an idea about the determinants and potentially about the of the future level of rent and yield pattern is very advantageous.

The author has experienced in the institutional real estate corporate market in Czech republic and Slovakia the fast growth in the mid-2000s’and the subsequent fall, followed by low liquidity and stagnation of this sector.

Most of the assets, rental contracts, and also liquidity and trades are concentrated in the office market in the capital city (Prague) [10] . In this thesis the author will thus concentrate on commercial office market in Prague.

Research Questions

Saunders (2007) defines research as collection and interpretation of information in order to find out things in a systematic way, thereby increasing their knowledge. Two phrases are important in this definition: ‘systematic way’ and ‘to find out things’. ‘Systematic’ suggests that research is based on logical relationships and not just beliefs (Ghauri and Grønhaug 2005). Systematic research involves an explanation of the methods used to collect the data, arguing why the results obtained are meaningful, and explaining any limitations that are associated with them. ‘To find out things’ suggests there are a multiplicity of possible

purposes for your research. These may include describing, explaining, understanding,

criticising and analysing (Ghauri and Grønhaug 2005).

Research question can be defined as a statement that identifies the phenomenon to be studied (Campbell et. al., 1982), or as stated by Saunders (2007), research question is one of a number of key questions that the research process will address. These are often the precursor of research objectives.).

Acording to Saunders (2007) one of the key criteria of your research success will be whether you have a set of clear conclusions drawn from the data you have collected. The extent to which you can do that will be determined largely by the clarity with which you have posed your initial Research questions.

Clough and Nutbrown (2002) use what they call the ‘Goldilocks test’ to decide if research questions are either ‘too big’, ‘too small’, ‘too hot’ or ‘just right’. Those that are too big need significant research funding and high level research, because they demand too many resources. Questions that are too small are likely to be of or trivial for a research thesis, while those that are too ‘hot’ may be too inadequate because of sensitivities that may be aroused as a result of doing the research [11] . Research questions that are ‘just right’, are those that are ‘just right for investigation at this time, by this researcher in this setting’.

3.1 Meaning of Research

V Cano (2002) has stated that research is about finding new things by evaluating

the work that is done by others in the past or present. In other words, research is a

systematic approach of finding results and solutions for the undertaken problem.

Saunders et al (2007) has stated that research means "Re-search", means ‘search again’ to

get better or different results for the undertaken phenomenon. Waltz and Bausell (19811)

have defined research as "a systematic, formal rigorous and precise process employed to

gain solutions to problems and/or to discover and interpret new facts and relationships.

Payton (19794) has stated that research is "a process of looking for or finding out a

specific answer to a specific question in an organized objective reliable way". V Cano

(2002) has stated that research can be used as an instrument to test an existing theory or

to create a new theory because it is a method through which the data would be collected

and analyzed to test or create theory.

What drives the prime capitalisation yield in of office transactions?

What will be the prime capitalisation yield in of office transactions in the following years?

What drives the prime office rent levels?

What will be the prime office rent levels in the following years?

Research Objectives

Formulating and clarifying the research topic is the starting point of your research project

(Ghauri and Grønhaug 2005; Smith and Dainty 1991).

The issues within the research are capable of being linked to theory (Raimond 1993).

Theory is defined by Gill and Johnson (2002:229) as ‘a formulation regarding

the cause and effect relationships between two or more variables, which may or may

not have been tested’.

Maylor and Blackmon suggest that such personal objectives would be better were they

to pass the well-known SMART test. That is that the objectives are:

• Specific. What precisely do you hope to achieve from undertaking the research?

• Measurable. What measures will you use to determine whether you have achieved

your objectives? (e.g. secured a career-level first job in software design).

• Achievable. Are the targets you have set for yourself achievable given all the possible

constraints?

Realistic. Given all the other demands upon your time, will you have the time and

energy to complete the research on time?

• Timely. Will you have time to accomplish all your objectives in the time frame you

have set?

Research objectives

(Saunders 2007) research objectives Clear, specific statements that identify

what the researcher wishes to accomplish as a result of

doing the research.

Saunders et al (2007) has defined research objectives "as a clear specific

statements that identify what the researcher wishes to accomplish as a result of doing

research". Further, they have stated that research objectives are the evidence which

represents the researcher’s clear sense of purpose and direction of the research study to

the community or readers. Maylor and Blackmon (2005) have mentioned that the

researcher can add his or her personnel objectives with the list of original research

objectives of their study. These personnel objectives could be the researcher’s learning

objectives and reflections from completion of research.

R F Taflinger (1996 & 2011) has stated, basically there are two purposes for any

kind research study and they are ‘to learn something’ or ‘to gather evidence’. The first

purpose is to learn something which would help the researcher to gain new knowledge

about something through the study or during the research study as learning is the ‘never

ending’ process. The second purpose is to use this learning as an evidence to show other

that what you have learned is correct. But, this learning and gaining evidence depends on

the researcher’s ability, interest and dedication towards the research study and also clear

research questions and objectives have its role in it. As said by R F Taflinger (1996 &

2011), the researcher of this present study have two purposes, the first one is to gain

better understanding about the importance of customer service and its impact on customer

satisfaction. Second purpose is to identify and analyse the impact of customer satisfaction

on customer loyalty in Starbucks, UK.

DEFINITIONS OF RESEARCH DESIGN:(1) According to David J. Luck and Ronald S. Rubin,

"A research design is thedetermination and statement of the general research approach or strategy adopted/or the particular project. It is the heart of planning. If the design adheres to the researchobjective, it will ensure that the client's needs will be served."

(2) According to Kerlinger

"Research design in the plan, structure and strategy of investigation conceived so as to obtain answers to research questions and to control variance."

(3)

According to Green and Tull

"A research design is the specification of methods and procedures for acquiring the information needed. It is the over-all operational patternor framework of the project that stipulates what information is to be collected fromwhich source by what procedures."

The overall aim of the diploma thesis is to analyze which macro and micro economic

factors influence yield rates for office properties in Prague. This goal will be

achieved by the analysis of the economic and real estate environment in relevant markets.

Analyse macroeconomic and real estate data with the goal to find correlations and causalities that drive the capitalisation yield and the rent levels.

Collected rata will be used to formulate an econometric model and to prove or disprove following hypotheses.

1. Yield rates are influenced by: yields in major markets (London), prime rents, local GDP, vacancy rates, and take-up, …

2. Prime rents levels are influenced by local GDP, vacancy rates, and take-up

Certain independent variables can have impact on the Yields directly others through the prime rents.

Literature review

Two major reasons exist for reviewing the literature (Sharp et al. 2002). The first, the preliminary search that helps you to generate and refine your research ideas, has already been discussed in Section 2.3. The second, often referred to as the critical review or critical literature review, is part of your research project proper.

real estate

Definition of

commercial real estate

Definition of

office real estate

Definition of

+ rozdeleni

A study by Hendershott and MacGregor concludes (2005) that UK office and retail property investors behave rationally. But they also find in another study that Australian and US office property investors do not (2005).

Definition of yield/capitalisation rate

+formulas

Determinats of the yield

According to Ball & Lizieri & MacGregor (1998) the capitalization rate is considered to be a function of risk free rate, risk premium, anticipated growth of rent and depreciation rate.

There are several studies examining the determinants of the yields. Froland (1987), Ambrose & Nourse (1993), Sivitanides & Southard & Torto & Wheaton (2001) show different connections between capitalization rates and financial, real economy and specific real estate characteristics factors like vacancy rate, completion rate and absorption rate. All these studies work with datasets of mature real estate markets in the USA. D’Argensio & Laurin (2008) investigated the determinants of office capitalization rate on panel of 52 countries (developed and emerging) in period 2000 – 2006. They found the 10 year government bond yield being the main determinant of the office capitalization rate. In this paper we focus on financial and macroeconomic factors influencing real estate (yields) in the Czech Republic.

Jun chen, Susan Hudson-Wilson, and Hans Nordby (2004) examine the connextions between capital amekrts and the real estate markets, and the relative investment enviroment using cap rate spreads. or tisk premium, between the implies cap rate and the 10-year Treasury bond, or risk free rate. The study observes the historical real estate performance and prcing cinfducts quantitative analysis of the pricing of the four property types- office, appartment, retailm, and industrial. the eplanatory variables used in the modelling are lagged cap rate spread, change of cosntruction, rent growthm rent cycle, vacancy change, inflation rate, gdp growth, S&P 500 stoch index growth. [Jun Chen, Susan Hudson-Wilson, Hans Nordby, 2004, "Real Estate Pricing: Spreads and Sensibilities—Why Real Estate Pricing Is Rational?]

The aim of the literature review is to:

[1] be familiarized with the real estate market, [2] to provide sufficient knowledge

background for yield forecasting, [3] to summarize the main economic factors

influencing development of the commercial real estate market in Prague

and [4] to familiarize with the main terminology of econometric modeling.

The most influential demand drivers for real estate properties according to McMahan are: economics, technology, and population and demographics. Aggregated demand

and supply of both commercial and private real estate property is a result of interaction among these factors. Further research conducted by Hartzell, Eichholtz and Selender (1993) pointed out the importance of additional economic driver -

employment rate. [2] [4]

Market demand drivers can be divided into two categories: a) driven by households, and b) driven by firms. These two categories are connected through employment, which generates

profit for firm and income for households. [4]

Time series are defined as a set of observation from past to the present time. Distance

between observations has to remain constant through all observations. Stationary

time series are defined as time series which main characteristic do not change

between observations. Non-stationary time series change through time. Two main

approaches toward time series observations are statistical and econometrical.

Econometrics

Exact interpretation of the term econometrics is economic measurement.

Econometrics describes key economic forces by using mathematical and statistical

methods. Real estate econometrics can be defined as application of statistical data to

real estate market, for examining changed in economic conditions and its linkages to

real estate market. Picture 10 explains steps involved in an econometric model. [32]

[33]

Source: Real estate modeling and forecasting, [33]

Step 1a and 1b – Formulation of econometric model

In this phase is usually formulated theoretical model. Model should be related to real

estate theory and should describe the linkage between variables. Models always

simplify reality and therefore cannot depict all real-world phenomenon, but it should

be able to present sufficient approximation. [32] [33]

Step 2 Collection of data

Collection of sufficient amount of data is crucial part of econometric model. Real

estate data can be collected from national and supra national statistical offices,

directly from firm or even from specialized researcher agencies.

Empirical analysis may use three types of data:

Time series data is a sequence of observation of a given variable at different times.

The interval between each measurement should remain consistent. Different variables

use different frequency of observations.

Cross section data are observations of different variables at the same point of time.

An example of cross section data is census of population. At the year of the census,

statistical office collects many variables.

Pooled data is a combination of both mentioned approaches. Example of pooled data

may be surveys conducted by Eurostat. It collects data from EU member states (cross

sectional data) and compares the results between years (time series data). [32] [33]

Elementary Characteristics of Time Series

Elementary characteristics of time series analysis consist of first (second, etc.)

differences and/or growth rate coefficient. Firs difference in a time series is defined as

series of changes from one period to next.

Formula:

Growth rate coefficient is a ratio of the value at current observation, divided by value

from previous observation. [29]

Formula:

Step 3 Model estimation

Economic model enables mathematical and statistical formulation of economic

theories. Econometric model broadens the economic model by adding stochastic

variable – error term. [32] [33]

Step 4 Statistical verification of parameters

Statistical verification of estimated parameters is necessary for testing the significance

of parameters. [33]

Types of variable

Variables are commonly referred as dependent and explanatory, or other. [32]

Given by a number of explanatory variables, models are referred to as simple or twovariable

(if we study dependence of one explanatory variable), multiple regression

analysis study more than one explanatory variable.

Special type of variable which needs to be present in every econometric model is

stochastic variable, often called error term. It consists of missing variables which may

influence the model, errors in measurement, and errors cause by wrong function form.

[34]

Theoretical framework

In this chapter the main terms and micro economical relationships will be discussed which are eventually necessary to understand the developments and characteristics of capitalisation yields and rents

Outline of research methods and sampling techniques to be used

SAUNDERS 2007 =

data Facts, opinions and statistics that have been collected together and recorded for reference or for analysis.

methods refer to The techniques and procedures used to obtain and

analyse research data, including for example questionnaires,

observation, interviews, and statistical and non-statistical

techniques. (Saunders 200?)

data that have already been collected for some other purpose. Such data are known as secondary data. Most automatically think in terms of collecting new (primary) data specifically for that purpose.

Many authors draw a distinction between qualitative and quantitative research (e.g. Bryman, 1988; Easterby-Smith et al. 2008). These are helpful in terms of understanding what is necessary in order to be able to analyse these data meaningfully.

Quantitative data

Qualitative data

Based on meanings derived from numbers

Based on meanings expressed through words

Collection results in numerical and

standardised data

Collection results in non-standardised data

requiring classification into categories

Analysis conducted through the use

of diagrams and statistics

Analysis conducted through the use of

conceptualisation

Based on Dey (1993); Healey and Rawlinson (1994); Saunders (2007)

Quantitative Research Methods

The functional or positivist paradigm that guides the quantitative mode of inquiry is based on the assumption that social reality has an objective ontological structure and that individuals are responding agents to this objective environment (Morgan & Smircich, 1980). Quantitative research involves counting and measuring of events and performing the statistical analysis of a body of numerical data (Smith, 1988). The assumption behind the positivist paradigm is that there is an objective truth existing in the world that can be measured and explained scientifically. The main concerns of the quantitative paradigm are that measurement is reliable, valid, and generalizable in its clear prediction of cause and effect (Cassell & Symon, 1994).

Being deductive and particularistic, quantitative research is based upon formulating the research hypotheses and verifying them empirically on a specific set of data (Frankfort-Nachmias & Nachmias, 1992). Scientific hypotheses are value-free; the researcher's own values, biases, and subjective preferences have no place in the quantitative approach. Researchers can view the communication process as concrete and tangible and can analyze it without contacting actual people involved in communication (Ting-Toomey, 1984).

The advantages of quantitative methods are mong oter the following:

Stating the research problem in very specific and set terms (Frankfort-Nachmias & Nachmias, 1992);

Clearly and precisely specifying both the independent and the dependent variables under investigation; (Alexei V. Matveev, 2002)

Following firmly the original set of research goals, arriving at more objective conclusions, testing hypothesis, determining the issues of causality; (Alexei V. Matveev, 2002)

Achieving high levels of reliability of gathered data due to controlled observations, laboratory experiments, mass surveys, or other form of research manipulations (Balsley, 1970);

Eliminating or minimizing subjectivity of judgment (Kealey & Protheroe, 1996);

Allowing for longitudinal measures of subsequent performance of research subjects. (Alexei V. Matveev, 2002)

Based on Saunders and Alexei 2012, quantitative methods are to be used eg. in case of: cause and effect reseseach, testing hypothesis and examinig relationhips, project and forecast results for a population, etc.

Considering the fact that quantitative data is available and there is high probability of them being reliable, and especially considering research question and the research objectives - this paper is using a quantitative research methods on quantitative data.

Quantitative data in a raw form, that is, before these data have been processed and analysed, convey very little meaning to most people. These data, therefore, need to be processed to make them useful, that is, to turn them into information. Quantitative analysis techniques such as graphs, charts and statistics allow us to do this; helping us to explore, present, describe and examine relationships and trends within our data.

Virtually any business and management research you undertake is likely to involve some numerical data or contain data that could usefully be quantified to help you answer your research question(s) and to meet your objectives. Quantitative data refer to all such data and can be a product of all research strategies

regression analysis The process of calculating a

regression coefficient and regression equation using one

independent variable and one dependent variable. For

data collected from a sample, there is also a need to

calculate the probability of the regression coefficient

having occurred by chance alone. See also multiple

regression analysis, regression coefficient, regression

equation.

regression equation Equation used to predict the

values of a dependent variable given the values of one or

more independent variables. The associated regression

coefficient provides an indication of how good a predictor

the regression equation is likely to be. See regression

coefficient.

regression coefficient Number between 0 and _1

that enables the strength of the relationship between

a numerical dependent variable and a numerical

independent variable to be assessed. The coefficient

represents the proportion of the variation in the

dependent variable that can be explained statistically by

the independent variable. A value of 1 means that all the

variation in the dependent variable can be explained

statistically by the independent variable. A value of 0

means that none of the variation in the dependent

variable can be explained by the independent variable.

See also regression analysis.

multiple regression analysis The process of calculating

a coefficient of multiple determination and regression

equation using two or more independent variables and

one dependent variable. For data collected from a sample,

there is also a need to calculate the probability of the

regression coefficient having occurred by chance alone.

See also multiple regression coefficient, regression analysis,

regression equation.

multiple regression coefficient Number between

0 and _1 that enables the strength of the relationship

between a numerical dependent variable and two or

more numerical independent variables to be assessed.

The coefficient represents the proportion of the

variation in the dependent variable that can be

explained statistically by the independent variables. A

value of 1 means that all the variation in the dependent

variable can be explained statistically by the independent

variables. A value of 0 means that none of the variation

in the dependent variable can be explained by the

independent variables. See also multiple regression

analysis.

dependent variable Variable that changes in response to changes in other variables.

independent variable Variable that causes changes to a dependent variable or variables.

negative correlation Relationship between two variables for which, as the values of one variable increase, the values of the other variable decrease.

positive correlation Relationship between two variables for which, as the value of one variable increases, the values of the other variable also increase.

chi square test Statistical test to determine the probability

(likelihood) that two categorical data variables are

associated. A common use is to discover whether there are

statistically significant differences between the observed

frequencies and the expected frequencies of two variables

presented in a cross-tabulation.

correlation The extent to which two variables are related

to each other. See also correlation coefficient, negative

correlation, positive correlation.

correlation coefficient Number between _1 and

_1 representing the strength of the relationship between

two ranked or numerical variables. A value of

_1 represents a perfect positive correlation. A value

of _1 represents a perfect negative correlation. Correlation

coefficients between _1 and _1 represent weaker positive

and negative correlations, a value of 0 meaning the

variables are perfectly independent. See also negative

correlation, Pearson’s product moment correlation

coefficient, positive correlation, Spearman’s rank

correlation coefficient.

chi square, Cramer’s V and phi to test whether two variables are significantly associated; - correlation

chi square test Statistical test to determine the probability

(likelihood) that two categorical data variables are

associated. A common use is to discover whether there are

statistically significant differences between the observed

frequencies and the expected frequencies of two variables

presented in a cross-tabulation.

To assess the strength

of a relationship

between one

dependent and one

independent variable –Coefficient of determination (regression coefficient)

To assess the strength

of a relationship

between one

dependent and two or

more independent

variables - Coefficient of multiple determination (multiple regression coefficient)

To predict the value of

a dependent variable

from one or more

independent variables - Regression equation (regression analysis)

Econometric analysis

approach is based on variety of standard Ordinary Least Square models (OLS). In that section we examine the relations between office yields and macroeconomic variables. In the OLS models we identify the most suitable model which would have the highest predictive power and would be in compliance with the described theory.

Given by a number of explanatory variables, models are referred to as simple or twovariable

(if we study dependence of one explanatory variable), multiple regression

analysis study more than one explanatory variable.

Special type of variable which needs to be present in every econometric model is

stochastic variable, often called error term. It consists of missing variables which may

influence the model, errors in measurement, and errors cause by wrong function form.

[34]

Dependent variable(YY) , explanatory variable (YY XX?)

Regression analysis

Term regression was first introduced by Francis Galton in the 19th century in his

research paper evaluating height of children based on the height of their parents.

Modern regression analysis is defined as dependence of one variable (the dependent

variable) on one or more other variables (explanatory variables). Example of regression

is depicted on Chart 8. Scatter points are actual measurements and curve depicts the

regression line. Theoretical points based on regression analysis. Main aim of regression analysis is to identify statistical dependence of one dependent

variable on one or more explanatory variables. Once statistical dependence is

identified, and goal is to estimate and predict the mean (average) value of dependent

variable. [32]

Ordinary Least Square Method (OLS)

The principle of ordinary least squares method is fitting a line to the data by

minimizing the sum of squared residuals as depicted on below chart. For regression

model the OLS estimators are BLUE (Best, Linear, Unbiased, Estimator) when:

The regression is linear in the coefficients, it is correctly specified and has an

additive error term.

Mean of the error term is zero.

The independent variables are not correlated with the error term

Residuals are not correlated with each other.

The error term has a constant variance

Multi-collinearity does not occur.

Causation

Strong correlation proved by statistics does not imply causation. Causation cannot be

predicted only by statistics it generally requires proof by other theories. For example

statistics can describe strong regression between yields of crop and temperature, but

other theories are required to describe and assess this relationship. [32]

Correlation analysis

The main objective of correlation analysis is to measure the strength or degree of

linear connection between two variables. Regression analysis treats differently

dependent and explanatory variables. Dependent variable is assumed to be statistical

and randomly distributed, explanatory variables are on the other hand assumed to

have fixed values. Correlation analysis does not distinguish between those two types

do variables, both are assumed to be randomly distributed. [32] [33]

Data analysis techniques to be used

Data validity and reliability

In the models varios data souirces will be used.

Public free data from statistical agencies

Public free and paid data from agencies – JLL, CBRE, C&W, ..

Public paid data sources form agencies like Bloomberg / Reuters

Public paid data sources from specialised databank agencies like – IPD,. … …

All data should be reliable to the level of the quality and reliability of the source agency. All agencies are global reputable and well known and reliable.

Data can have standard statistical noise and not be 100% precise as it depend on the methodology of the relevant institutions. Methodologies in case of statistical agencies and Eurostat are public, in case of private provides data collection methodology is often provided with the paid services.

Ethical issues

Saunders 2007 - research ethics The appropriateness of the researcher’s

behaviour in relation to the rights of those who become

the subject of a research project, or who are affected by it.

Qualitative research is likely to lead to a greater range of ethical concerns in comparison with

quantitative research, although all research methods have specific ethical issues associated

with them.

Ethical issues need to be considered in…

In the models various data sources will be used. As mentioned before in Data validity and reliability, the data used will be from public paid and unpaid sources.

At least in one of the agencies that provided data requested a proclamation that the data will not be provided for direct further commercial use (resale of data).

No personal information of natural persons will be used, and data and results published will not be in conflict with the various laws for the protection of personal information in CR (or other relevant jurisdictions)

No private data will be used from company of my employer.

No data will be used that is confidential or would be in conflict with the various insider information restriction laws in CR (or other relevant jurisdictions).

General R/E market information, R/E standards and R/E experience of the author are blending throughout the thesis.

Research contribution

Since in the revolutionary paper Case and Shiller (1989) demonstrated that the US real estate housing market is not informationally efficient and thus does not follow a random walk, the execution of forecasts in real estate is proven to be possible, reasonable and had been demanded. Their finding was explained by the heterogeneity, illiquidity, and high transaction costs that underlie all real estate markets. In case a market is not efficient, information is not equally spread and thus analysis of the trends, market patterns, causalities and consequent forecasts about the future can yield a competitive advantage over the other market players (either competition or the counterparties).

Who can benefit from the research results? For short and mid-term term investors and developers the understanding of causalities forming the prices (correlations, regressions) and having forecasts based on certain assumptions can be crucial (ie yield forecasts), for mid to long term investors the same applies bud additionally the understanding of the rent levels is crucial, as the rent levels can influence the prices more than the multiples (yields) can. For tenants the understanding of the rent levels is crucial as well. Especially in case their property leasing costs represent such a level of their total expenses that it is worth a wile considering the future rent and price development. They can speculate whether it is more economical to lease a property or to buy a property. (in case their return on assets is higher than the net yield)

When is it more economical to buy a commercial property? - This happens in case company’s return on assets is lower than the net yield of the considered property for the selected period (in case leverage is used, the leverage and interest rate needs to be included in the model). This is usual in situations of high real estate yields and low growth and decreasing or stagnating productivity – ie. in recessions and stagnations.

When is it more economical to lease or back lease a commercial property? - This on the contrary happens in case the return on company assets is higher than the net yield of the considered property for the selected period (in case leverage is used, the leverage and interest rate needs to be included in the model). This is usual in situations of high liquidity (oversupply of capital) and low real estate yields, and high growth and increasing productivity – ie. in an expanding economy. [12] Tenants can also speculate on the length of the lease contract when they have expectations on the future developments of the rent (are the market rent about to increase or decrease?).

Institutions providing debt capital (lenders – banks and buyers of bonds) need to adjust their level and pricing based on the expected future value of the property, to have a safety buffer for the money the lend (Loan to value ration can and does change significantly through the time)

In any market the participant operates (space market, asset market, development market, financing market) it is favourable for him to understand the causalities and the future development of yields and rents.

The results and contribution of the rigorous thesis should be minimally as follows:

Proving and disproving relationships that drive the most important office R/E figures – i.e. yield and rent levels in Prague.

Creating a model for the forecasting of yield and rent levels.

Forecasting the figures for yield and rent levels for future periods based on the relationships discovered.

Project Plan

The project plan for the research and the dissertation thesis is as follows:

February – March 2013

analysis of the literature

analysis of various data sources

preliminary analysis of various econometrical and statistical SW

discussions with the data source owners about possible usage

first sample data collected

April – May 2013

data collection and data verification

additional literate review

additional review and deeper understanding of econometric and statistical models

additional study of statistical SW

testing the data

June 2013 - and later

calculating the correlations

calculating the regression model

forecasting the variables into the future

writing and finalising the rigorous thesis based on the data and results found

Bibliography (Harvard reference system)

Assignment criteria and marking scheme

The following criteria will be used to assess the completed proposal:

Title and Introduction (10%)

Clarity of title (accurate and concise)

Background information provides an overview of the topic to be investigated

Appropriateness of aim, research question or testable hypothesis

Literature Review (25%)

Literature selected is appropriate to the aim of the proposed study

There is critical analysis of the literature selected

Critical analysis is meaningful and of sufficient depth

Research Methodology (20%)

Selection of research design is justified

Indication of method of the chosen subject

Justification of chose method(s) of data analysis

Ethical Issues (10%)

Consideration of issues of confidentiality and data security

Indication of any risks to subjects and/or researcher

Research Plan (10%)

Clear, reasonable indication of a research strategy and structure

Conclusion (5%)

Indication of possible study limitations

Appropriate summary of the main emphasis of the research proposal

Bibliography (10%)