MedicalBiostatistics.com
Biostatistics Resource for Medical, Health and Allied Research
Home

Weather Forecast in India

Weather forecast is given prominence these days in newspapers and TV broadcasts. But forecast in India reminds me of what my son said when he was a sixth grader. If dry hot weather is predicted for all days of April, May and June, rainy for July, August and September, cold for the months of December, January and February, and mild for the remaining months, the forecast will be right for at least 70 percent days. Being a statistician, that did not surprise me at all. I am not a meteorologist but if I predict the same weather for tomorrow as is today, believe me I will be right 90 percent times! One does not need to be an expert to achieve a predictivity of 90 percent in Indian setup where the weather is so stable compared to many European countries or the US.

That is what our met office does. When the temperature shoots up, they will say that it is likely to continue for some more days. They rarely correctly predict the change in weather. They never predicted hazy conditions for Delhi, or for small pre-monsoon showers. They must realize that efficacy of weather forecasts in a country like India does not depend on stable days but on the variable days. If a forecast of a change by our met office really turns out to be correct, it is so more by chance than by design. This is like marking all ‘a’ answers in a multiple choice questions test by a kid, and scoring 25 percent marks. An inexpert like me knows that if the sky is overcast today, it is safe to bet for scattered showers tomorrow. If luck favours, I would be right. If not, such unscientific forecast will still have substantial chance of success in the long run.

It is time that our met office learns a lesson or two, and assess the efficacy of their prediction in a stable-weather country like India by their ability to predict change in weather conditions, and accordingly improve their techniques of prediction.

PS: Please ask the met. officers why they consistently fail to forecast a change in weather conditions.