Linear Prediction of Indian Monsoon Rainfall*

This paper proposes a strategy for selecting the best linear prediction model for Indian monsoon rainfall. In this strategy, a cross-validation procedure first screens out all models that perform poorly on independent data, then the error variance of every remaining model is compared to that of every other model to test whether the difference in error variances is statistically significant. This strategy is shown to produce better forecasts on average than selecting either the model that utilizes all predictors, the model that explains the most variance in the independent data, or the model with the most favorable statistic suggested by Mallow. All of the model selection criteria suggest that regression models based on two to three predictors produce better forecasts on average than regression models using a larger number of predictors. For the period up to 1967, the forecast strategy selected the prior climatology as the best predictor. For the period 1967 to the present, the strategy performed better than forecasts based on the prior climatology and all other methodologies investigated. Indexes of Pacific Ocean temperature in the Tropics (called Niño-3) and indexes of pressure fluctuations in the Northern Atlantic (called NAO), at 1–6 lead months, failed to provide prediction models that performed better on average than a prediction based on the antecedent climatology. Forecasts based on the prior 25-yr climatology had especially high skill during the recent period 1989–2000, a fact that appears to be a mere coincidence, but which may be relevant to interpreting the skill of the power regression model currently used by the India Meteorological Department.