While R2 is one parameter to look out for, the most standard way is to check the residual plotst to evaluate the model. Let's apply the regression technique and discover if the assumptions get validated without which our model doesn't stand a suitable fit for the given data and future predictions. Though it is not an exact linear relation, the whole purpose of modelling is to understand the uncertainty complemented with statistical analysis. Error terms should be normally distributed with mean 0.All independent variables are uncorrelated with the error term.Observations of the error term are uncorrelated with each other.There should not be any outliers present.There must be a linear relation between independent and dependent variables.
What do the trends say?Īll the scatter plots helps use decide to go for a linear regression modelĪssumptions to apply linear regression model: This is a transit demand data set and the dependent variable under consideration is Number of weekly riders and there are four independent variables whose predictability we want to know.
The higher the proportion, the better is the relationship between dependent and independent variable/s. R2 is R-squared value which is defined as the measure of proprortion of variance of dependent variable explained by the independent variable.Microsoft Excel's regression limits to linear regression analysis however one can try to fit with one independent variable or multiple independent variables.This analysis has a small dataset which explores the dataset from application point of view and I hope it would be a precursor to complex analysis using the same or different technique.
0 kommentar(er)
