The main objective of this research work is to determine the vector autoregressive _x000D_ models (VAR) that best predicts the unemployment rate, GDP and CPI in Peru. With the help _x000D_ of the R programming language applied in RStudio. The methodology of VAR models was used, _x000D_ where the analysis was consolidated in graphs, tests such as Dickey Fuller, Phillips Perron and _x000D_ the Kwiatkowski-Phillips-Schmidt-Shin (KPSS) test to determine if a time series is stationary _x000D_ around a trend, for the validation of the model it was incurred to apply the three basic tests _x000D_ through which a model must pass (VAR), the Jarque-Bera normality test, No Autocorrelation _x000D_ and the Heteroskedasticity test.In the process of searching for a VAR model, a residual _x000D_ normality problem was incurred since the Jarque-Bera test showed a p – value = 0.00, _x000D_ consequently, i...