The forecasts on inflation and output growth - Research Paper Example

Research report Introduction When general prices of goods and services in a country rise, the value of money drops. This in return leads to inflation. Inflation then in the long run leads to the decrease of purchasing power of a currency. The occurrence of inflation has led to many central banks around the world to operate on a monetary policy regime of inflation targeting. This helps the financial institutions to organize themselves in such a way that they can forecast more on the future of the economy in order for them to know the best way to curb it when it actually happens. This paper discusses the forecasts on inflation and output growth, the way it can be improved and its usefulness. Discussion What is the failure on the projection of inflation and growth forecasts? Inflation projection is the forecast by economists of the current economy based on real time data and the effects it will have in the future of a country. This is to mean that when a country’s central bank does not practice inflation projection then it will not be able to control the rates when the actual inflation occurs. The failure in the other hand will lead to unclear and untimely communication of policy objectives, plans and tactics (Mandal, Gavin, & Mandal 2000). It can also lead to increase in economic and financial uncertainty (Bergheim 2008). Growth forecasts in the other hand are the prediction of a future economic growth which can be seen through the current economic times. Growth predictions are important as it prepares a country for an incoming increase in a country’s GDP. The failure of forecasting growth forecast is that a country will not be amply prepared for the increase and it will lead to poor budgeting (De Jong 2007). What “suite of statistical forecasting models’ to be used by the Bank of England? The bank of England has constructed models with which it can provide judgement free statistical forecast on the inflation and output growth. The suite focuses on combining in an optimal way a small number of forecasts generated using different sources of information and methodologies (Mandal, Gavin, & Mandal 2000). The models used are classified into three different categories. They are: 1. Benchmark forecast, 2. Forecasts from linear vector-auto regressions and 3. Forecasts from univariate non-linear models (Bergheim 2008). Benchmark forecast The first benchmark model is the suite is that the variable of interest is equal to the unconditional mean over the recent past. Another example is the random walk (RW) or no change. These models are often found to forecast very well. Auto regression is the third benchmark suite. In this model unvariate representations can often be captured by low order systems. While not robust to structural change they are robust to misspecification following incorrect choice of explanatory variables (De Jong 2007). Forecasts from linear vector-auto regressions Linear vector autogressions (VARs) are linear relationships between small set of variables. Economic interests in them relate to identification (structural VARs, SVARs) but unless the restrictions over identify the model which is not the case usually. In it includes basic VARs which are the standard linear reduced form of VAR (Bergheim 2008). There are also the double differenced VARs, forecast from Bayesian VARs; this form of VAR does not suffer from parameterisation. It uses priors with small number of parameters to extract the signal parsimoniously. The last one in the linear vector autogressions is the recursively-estimated VARs (Hudson 2006). Forecasts from univariate non-linear models In this there are a large number of non-linear models that are used to model unvariate processes. It includes models from the smooth transition autogressive (STARs), the threshold autogressive (TAR) and Markov Switching (MS) (De Jong 2007). Smooth transition autogressive (STARs) Estimation of the model is by non-linear least squares. The parameters are sometimes difficult to obtain, in which case a grid search may be used to obtain some ideas on their values before using non-linear least squares with the outcome of the grid search as initial conditions (Bergheim 2008). Markov Switching (MS) This model essentially implies that there are m regimes in the economy regulated by an unobserved Markov chain. The model can be estimated via maximum likelihood using the filter. The (MS) model in the suite uses two regimes for both mean and vitality (Hudson 2006). What best models explain the macro-economic stability in the Bank of England? The macro economic stability of the bank of England is best explained by the Benchmark forecast and Forecasts from linear vector-auto regressions. In the benchmark suite the factor of macro economic study is seen through the fact that the well-documented move towards macroeconomic stability, sometimes referred to as the ‘Great Stability has made forecasting more easy in the sense that macroeconomic variables stray less far from the unconditional mean than in the past; but more difficult in the sense that it is hard to outperform naive models (Nicholson 2010). Which is also the cases in Forecasts from linear vector-autoregressions only that this time round there are restrictions which identify the model which is not the case usually (De Jong 2007). Conclusion From the foregoing literature it can be seen that inflation and growth forecasts are vital for the stability of a country in the future. It is therefore important that a country invests in such opportunity. This will increase the country’s success and financial strength. This importance has led to the central banks of various countries to invest in the forecast information sector. Such investments it have enabled countries like Canada, New Zealand and Australia turn out to be economic giants. Reference List Bergheim, S. 2008, Long-run growth forecasting. Ohio: Springer. De Jong, P. J. 2007, The change in financial analysts' forecast attributes for value and growth stocks. London: ProQuest. Hudson, J. 2006, Inflation: a theoretical survey and synthesis. London: Routledge. Mandal, R. R., Gavin, W. T., & Mandal, R. T. 2000, Forecasting inflation and growth: do private forecasts match those of policymakers? St Louis: Federal Reserve Bank of St. Louis. Nicholson, S. 2010, Inflation. India: Nabu Press. Read More