Regression Analysis

Regression analysis is most commonly used in forecasting.  When running a regression analysis it helps to understand what you are trying to predict.  As an example when you want to predict the price of a comodity for the next 2 years.  Understanding the type of variables that correlate to what you are trying to forecast will help in the regression.  As an example if you want to generate a simple regression analysis to predict the copper price for the next 2 years.  Some variables that will correlate directly to copper price are gold, dow, gdp, Niikai Index, silver price.  Highly correlated variables such as these can be used to put together a mathematical formual to predict cooper price.  A regression starts with data so once you understand the variables that correlate to copper price you can pull forecasted numbers for those variables and use that to run a regression for copper price.  Tools like Excel can be used for forecasting and will make life a lot easier than trying to calculate it by hand.  One thing to remember is that forecasts are always wrong.  Expecially in a economy such as this one with constant unpredictability and market shifts and shorter forecasts are more accurate than longer forecasts.  Meaning that a 6 month forecast will be more accurate than a 10 year forecast.  In this economy creating a shorter forecast makes more sense.   Expecially for technology or commodity prices.