Regression is used to analyze the situation among variables; this is done by estimating a relationship. Regression can be simple with one independent variable (x), multiple which has more than one variable; in this model there are multiple regression model that are characterized with a k variable that is independent. Linear where the y variable are with parameters that are to be estimated and one value in an equation not an error item. For example the general use of indicators of economy, for instance the Dow Jones consumer price index and the industrial average that a producer or a suppler can use to forecast demand for their goods and services. These two variables can be used can be used to predict demand and also find out the variation of demand is as a result of the two variables (Yan, Su & World Scientific, 2009).
Regression analysis used to forecast by using time as a variable that is independent. For instance in an assumption that one’s revenues tend to increase linearly over time, in this case the year can be used as the x variables and the y variables can be the revenues. By doing so one can be able to predict how the variables will be for the whole year, as long as they do not go far from the present. Other things that can be forecasted using this formula are the price of the commodity as well as the number of suppliers that are available (Berk, 2004).
Making of individual prediction
If the variables of several explanatory for an individual, but the value of the individual dependent variable is not known, one can be able to estimate the value of the variables that re dependent of that individual by using prediction equation.
In connection to the statement above, one is able to predict maintainers and repair expense for a particular one year old Mercedes is in the motor pool at the moment, the first thing to be done is to perform by using the age and making the variables that are explanatory.
Cost p = 8.53 Age= 705.66-5427make.
According to the prediction then one can say that $714.19 and the 95% (margin of error) for this particular prediction is 124.0141 x 2.1788 = 270.20.
Effects of explanatory variable and dependent variable
It is not easy to have an estimation that is “pure” when it comes to explanatory variables. So as to try and make the prediction as accurate as possible, many other effects have to be controlled as much as possible. This would enable one to see the change of the prediction of a certain individual if the explanatory was to be made differently, as the same time all the other aspect of the prediction were kept the same. This can only be achieved if the most complete model that is available is used. All other factors should also be inclusive as additional explanatory variables.
According to the above statement, one can use this formula in the example below; by using a full regression model, it can be estimated that the mean marginal maintenance cost with diving one vehicle, in an addition of 1000 miles is $26.65, this can be inclusive of a margin error of 3.915 x 2.2010 = $8.62. for better understanding on the usage of complete model that is available, keeping in mind that any vehicle has a particular age and make as well as holding those constant considering the incremental effect of driving another 1000 miles.
References
Berk, R. A. (2004). Regression analysis: A constructive critique. London [u.a: Sage Publ.
Yan, X., Su, X., & World Scientific (Firm). (2009). Linear regression analysis: Theory and computing. Singapore: World Scientific Pub. Co.
