Indra Dev and monsoon forecasting: What is the dummy variable?

A super fun filled column by Manasi Phadke (of the crazy blog – Manasiecon.

Its starts with this scene where Lord Indra is being reported on rains in India:

The scene opens in the Indra Sabha, where LOTR, Lord of the, errr, ‘Rains’, presides in full glory over humbler portfolios like Costumes and Jewellery.

Sevak:My Lord, the Indian Meteorological Department has given its long range forecast. Its website claims the monsoon will arrive on time and hit Kerala on June 1.

Indra Dev (interested): Oh ho! Really? Tell me, Sevak, on what basis do they make such claims?

Sevak: Sir, the poor earthlings at the IMD still use statistical modelling for monsoon forecasting. Can you believe it, Sire? I mean, this “identification of predictors to forecast the predict, assuming that the nature of the long-run relationship between dependent and independents is stable” is boring ole OLS stuff. Almost Muggle stuff, you know, Sir! But Indian blizzard, being more wizard, defies predictability and, hence, prediction. If you’ll permit your humble servant a quick spell, Sir… Cumulonimbus cyclonum!

Indra: Confound that Potter woman! Not one of my trusted Sevaks can create a cumulonimbus in Sanskrit anymore. Sevak! Stop rolling about Rowling. And get Tolkien, I mean… talking.

Sevak: Ahm, yes, Sire! The earthlings, though ignorant and cocky, are scientific, you can’t deny that! The IMD indicates the model has held stably, when things have been, errr, stable. The current statistical model apparently goes into doubt only in case of a drought. 2002 and 2009, Sir! Those two years when you, for some strange reason, got upset…

Indra: Ahm, yes. What else do they have to say?

Sevak: They further claim that the error forecast on monsoons has never been more than 10 per cent in monsoon forecasting, Sir! And that seems to be the globally acceptable error limit!

Indra: Yeah, I did hear Zeus say the Greeks have started accepting really high limits these days. But I wonder if he was talking of monsoon errors, or simply debt levels. And how about China?

Sevak: Ah Sir, the Chi Chongzi! Such a temperamental Rain God, Sir! Makes Poseidon seem relatively harmless, Sir! Right through the ages, Chi has played a high variation game. The Tang, Yuan and Ming dynasties all experienced unrests and declines when Chi got upset. The Chinese earthlings, bless them, have written a paper on the ‘Effect of Climate Change on Dynasties’. So 10 per cent Chi-error margin for the Chinese is factored in, Sir. Just see their statistical models, Sire. Despite being ‘t’ fans, they use the ‘Chi-sq!uare’…

Indra: Ha! Statistics to predict monsoons! The Indian earthlings are close, really close. They regress monsoon probabilities on cloud formation, humidity, evaporation and other weather variables. But they are completely missing the most important variable! The dummy variable!

Sevak: Which is?

Indra: The Finance Minister. I like your FM and you are through.

Sevak (wonderstruck): Oh! Sire, what kind of FMs do you like?

Indra: I like the firang educated ones, Sevak! Give me a Cambridge or a Harvard graduate, and your rains are guaranteed! I never troubled good old Manmohan Singh or P Chidambaram, did I? I was so sharif with the kharif. Yashwant Sinha realised it in the nick of time and packed off Jayant to Harvard. But Jaswant and Pranab. No firangdegree, no rain. Had NaMo opted for Subramanian Swamy, I would have showered rains galore! But now, with Arun, I am not so sure.

Sevak: But after that late start, you did send NaMo a few clouds last year.

Indra: That’s because I’ve a soft corner for NaMo. Kind of namesake, you know. Nar-Indra!


Why just the FM? All policymakers should come from La La Land. Then all possible Lords bless India. Even our lords seem to be having a huge liking for foreign lords…sorry degrees..

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: