Mahmood Ul Hassan. Photo: Department of Statistics
Mahmood Ul Hassan. Photo: Department of Statistics

A country’s income, or Gross Domestic product (GDP), is one of the most important and traditional measures of economic growth. Mahmood Ul-Hassan, PhD scholar at the Department of Statistics, devoted his master thesis to find a more accurate way of forecasting GDP. Now, his findings are published in Journal of Applied Statistics.

- Everything planned is on base of the GDP growth. Because it indicates how rapidly the country is growing, says Mahmood Ul-Hassan.

In order to make a better prediction, you need to know the so called density function of the economic growth. The density function in statistics is used to describe how likely it is that you will get a certain result, compared with other results (the relative likelihood). In other words, and if you simplify it – by finding the right density function to describe economic growth, you will be able to predict the growth in a more accurate way.

Knowing economic uncertainty in advance

The starting point of Mahmood Ul Hassan’s research is this: economic uncertainty is the problem most of the planners worry about.  Economies all over the world often experience fluctuations totally unexpected; it goes up and down because of different ‘shocks’ This means, as Mahmood Ul Hassan describes it in his paper, that economic growth often shows irregular patterns both in the short and in the long run.

It also means that it is challenging to find the right density function to describe it. Mahmood Ul Hassan got the idea for his thesis and later paper, from his supervisor Pär Stockhammar who wanted to find a solution to this problem - and passed it on to him.

Data on the GDP-growth in three different countries were used in the study, seasonally adjusted series up to 2012 from the United States, United Kingdom and Canada.

Because of the asymmetry and other typical features (such as highly leptokurtic data and heteroscedasticity) of the data, Mahmood Ul Hassan found out that the so-called TAL2-distribution was the best match for his data from the US. For the United Kingdom and Canada, the so-called Normal Mixture, NM-distribution, was a better fit.

In addition to forecasting, the findings of Mahmood Ul Hassan’s study can be used to compare the economies of different countries.

Will know how much food to produce

There are many reasons to why you would want to predict how fast the economy is growing, for example when you make budgets and form strategies.

- If you can predict the GDP growth of a country, then the planners will be able make better and accurate strategy accordingly, says Mahmood Ul Hassan.