Econometrics: How to Predict House Prices Growth Essay Assignment

Assignment Task

Instructions

This is an essay, with arguments supported by empirical evidence. Hence, like any essay, it should have an ‘essay structure’ constituted of Introduction, Methodology, Findings, Ro- bustness Check and Conclusion.

• Introduction: brief motivation+background/literature. This part addresses why the topic is worth investigating.
• Methodology: theory, benchmark model (regression), data. This part is mainly to present and explain the chosen regression model. The choice should be based on reasonable supposed beliefs, such as good theories or relevant previous findings. It also explains the data used, including their sources, time range, measurement, and mostly importantly in time series analyses, whether they are stationary (or how the original data are processed such that the processed data are stationary).
• Findings: this part reports and interprets the findings of the benchmark regression. This should be the main focus, as it answers the central question of the essay. You can start with the estimation result, consider the estimated coefficient values including their size and sign; and evaluate their significance. Then you can comment on these results; e.g., what they mean, what the policy implications are and, (if your finding is different from what typical theories or previous author have found) how/why your finding is different from expected. (Note that there is only right or wrong method; there is no ‘right’ or ‘wrong’ result if the method is right. It would just be what the data says!)
• Robustness Check: this part checks ‘robustness’ of the finding established with your benchmark regression. You can check, e.g. whether the benchmark regression has problems of multicollinearity, heteroskedasticity or autocorrelation; or whether some of the RHS variables are endogenous so that an IV estimation would be better; or whether a dummy variable could be added to improve the regression’s fit (e.g. the Global Financial Crisis around 2008, the Covid pandemic since 2020Q1, or more recently the war in Ukraine). But it is impossible to consider everything. In practice, people would just consider one or two aspects. See Point 6 below for what are required for our essay.
• Conclusion: the brief round up of what you have done and

When reporting your regression result: create your own table and only report the relevant information (Do not copy the Eviews tables directly which would include too much irrelevant information). Using the Eviews figures is fine but you may like to use your own figure/variable names etc. Keep four decimal places.

Note this is an essay, not an Eviews assessment. You only use Eviews as a ‘calculator’. Therefore, Eviews screenshots should not be added to your essay.

What determines the house price growth in [Coun- try X]?

1. Choose a country you are interested in and investigate what factors affect the growth of house prices in that

2. Definition of house price growth: the percentage change in any general house price index of a country (An example of HPI can be found here For some countries a price index may not be available; but if there is data of house prices, e.g., xxx dollars/meter ² , then you can still calculate the price growth rate in the usual way. When percentage data is used in Eviews, use figures without the % sign; e.g. if the value is 5%, put 0.05 in Eviews.

3. Determine your own benchmark The LHS will be the house price growth (or call it house price inflation) anyway. The RHS factors are chosen by yourself. But you should consider a minimum of 3 factors in this benchmark regression.

4. Collect the data by yourself. Both annual and quarterly data can be used. But a minimum of 60 observations is required and, if quarterly data are used, they should not be older than

5. In your data session: use the ADF test (with a constant, but no trend) to test data stationarity. If any time series needed by the regression is found non- stationary, use the ‘linear detrending’ method taught in L7 to stationarise that time series (Tips: if cointegration exists in your benchmark regression, you don’t need to detrend the data as that would be the ‘special case’; b. if there is no cointegration and if linear detrending doesn’t help, consider measuring your variables in growth rate, as growth rates are normally stationary; c. if no luck with all the above, simply use another variable).

6. In your Robustness Check session: consider TWO of the following poten- tial issues (All should be compared to the benchmark regression): a) proportional heteroskedasticity (Select only one RHS variable to discuss), b) autocorrelation (Con- sider up to 4 lags for the error term), c) endogeneity of RHS variable (Select only one variable to discuss). For all these issues, consider how to test their existence and how to deal with the issues if they exist. In any case, discuss how your benchmark finding may be affected (i.e., whether it is robust to these issues).

7. All equations in your work should be A list of references (any style) should be provided.

8. Databases: use any free, publicly available sources you can find The several free resources used by many in the field of economics are: https://fred.stlouisfed.org/ (Mainly for US but also include data for main economies), https://www.ons.gov.uk/ (Mainly for UK), https://ec.europa.eu/eurostat/data/database (Mainly for European countries), https://data.oecd.org/ (Mainly for OECD countries), https://data.worldbank.org/ (Mainly for developing countries and emerging economies). You may also find data from the statistics department of the government of the selected country.

This Economics has been solved by our PhD Experts at My Uni Paper.