MS/M.Phil DQM (Economics and Qualitative Management

Permanent URI for this community

https://escholar.umt.edu.pk/handle/123456789/4093

Browse

Browsing MS/M.Phil DQM (Economics and Qualitative Management by Author "Fizzah Haq"

Now showing 1 - 1 of 1
  • No Thumbnail Available
    Item
    A Comparative Study of Spline Models
    (UMT.Lahore, 2020) Fizzah Haq
    Economic problems that faced in a country cause great changes particularly in the economy progression. Also they are influenced by various internal or external situations. Economic or political problems generally cause rapid and sudden increases in inflation and commodity prices. Gold has grown into an alternative investment mode for Pakistani investors. Price rise is defined as inflation, which is the increase in the cost of living as the price of goods and services increases. Inflation rate is the annual year percent change in price level. This report used the wholesale price index (WPI) as an inflation analog. It is commonly used by numerous analysts, states, banks and sectors, because it aims at price movements most comprehensively. The data of gold price and whole sale price index that used in this study was obtained from Census and Statistics Department State Bank of Pakistan from duration of 1960 to 2018, the sampling period consisted of 59 measurements per annum. The aim of this study is to compare two spline models regression spline and penalized spline and also to achieve a precise and accurate estimate for the price of gold. The prediction has been made for Pakistan's bases of inflation rates over the last six decades.Mean square error (MSE) and mean absolute percentage error (MAPE) were used to evaluation criteria to test the reliability of both analytical techniques. R-studio 3.5.2 is used for this imperial analysis. Best result hasbeen achieved using the cubic penalized spline according to MSE and MAPEperformance criterions. F-test is used to compare the two spline models and conclude that both techniques are nor statistically significant neither functionally relevant, both models are significantly different. The R^2 significance of the penalized spline model for gold price data is higher than the R^2 of regression spline model.Thus, it is conclude that specially in case of prediction the penalized spline model is a more effective and appropriate statistical method for gold price data and also for such types of economic data in Pakistan.