In the classical linear regression model we assume that successive values of the disturbance term are temporarily independent when observations are taken over time. When this assumption is violated then there is a problem known as Autocorrelation. In the presence of the problem of autocorrelation, the value of the standard error of the parameter estimates are affected and the predictions based on ordinary least square estimates will be inefficient. In this study main focus given on the different consequences of autocorrelation, detection of the problem of autocorrelation and also its different remedial measures. Consequences are explains both in case of two variable model and in case of k-variable model separately. To detect the problem of autocorrelation several tests like Durbin-Watson test, Theil-Nagar test, Durbin h-test, Breusch-Godfrey Lagrange Multiplier Test, and Run Test are used. We also explain different remedial measures to encounter the problem of autocorrelation. At last we consider a data set of three variables and explain the procedure of detection of autocorrelation and its remedial measure using STATA 14, econometric software.