Many years ago, when I was still living in New York, I remember waiting for the elevator in a building mid-town, when these two guys came in stomping snow from their boots. One of them turned to the other and said, "Y'know, weather forecasting is the business to get into. You get paid even if you're wrong."

I think of those guys quite frequently, and wonder whether I have missed my calling. Wandering through reams of data on wind speed and precipitation and humidity and God knows what else is probably a lot like trying to understand markets. And, likely, just as frustrating.

Which is why I am always suspicious of market analyses -- whether 'fundamentalist' or technical. Using scientific methodologies to try and analyse forces that are at-best science plus always -- being euphemistic -- leaves plenty of room for interpretation. The trick, of course, is in figuring out the interpretation.

Let me give you an example. Back in August 2007, when the rupee was around 40, with reserves rising at about $ 3 billion a week, and the RBI huffing and puffing to control the rupee's strength, our technical analyst came up with a forecast that the rupee would fall to between 45 and 46 to the dollar in October 2008. 

I asked him whether he had lost his mental balance. Capital flows were surging -- the July to September quarter saw inflows of $ 35 billion into Indian markets -- and the only way, it seemed, that the rupee could go was up and up and up. Indeed, 38 to the dollar was the talk of the town.

Our analyst -- and more power to him for that -- said, "Well, you know, if I use a different wave count, the rupee could weaken even further -- to below 47.50."

I buried the forecast, of course, and wondered what to do with him as the rupee powered through 40 in September 2007 and stayed in a tightly wound 39-40 band all the way till March 2008. 

By then, of course, the credit crisis had come home to roost, inflows had more or less completely dried up, nobody was lending to anybody else, and in India, as everywhere else, all hell broke loose. The rupee fell like a stone, and by Diwali this year, had smashed through our analyst's target of 47.50.

I couldn't believe it. I dusted off his report and, sure enough -- he had been correct. At least on direction and timing, if not on the degree of the decline. I kicked myself and wondered how I could have possibly convinced people to buy dollars back in 2007, when buying rupees seemed to be the only game in town.

Properly chastened, I called for his current forecast and here it is.

After hitting a high of 86.98, the dollar index has slipped a bit. This rise represents a rally of almost 22 per cent from its all-time low of 70.67, registered on March 17 this year. Judging from the wave structure, the rally now seems complete with a complex double zigzag (ABC-X-ABC). While a triple zigzag -- with another X-ABC to come -- cannot be ruled out, the DXY should make only marginal gains beyond the current high and that too after a deep cut (X) to around 80. We believe the DXY's bottom in the current cycle is still ahead of us and we expect it to bottom around 70 (a little below the March low) in early 2010, implying a decline of 19-20% from the 86.96 high.  

The rupee should closely track the DXY, but with a slightly weaker bias given India's twin deficits and the forthcoming elections. This would imply a top for USD/ INR around 50 and a maximum decline to 40 (20% gain). There is a clear 5-wave pattern visible in the rise of this year, with the 5th wave being hyper-extended. In the worst case, we could just be completing the 3rd of 3rd wave, with two more upward thrusts corresponding to the A and C of the potential third zigzag in DXY. That would imply a floor of 46.50/47 for the next leg and a top around 52-53 a few months down the line. The target for the subsequent major decline would have to be raised proportionately to 42.

The good news is that this isn't anywhere near as outlandish as the one last year. It isn't hard to understand a weakening dollar -- once the world gets over the liquidity terror, and with LIBOR down near 2.5 per cent we are almost there, it will become clear that a nearly 2 trillion deficit is bound to lead to a weaker dollar. A stronger rupee, too, seems reasonable, given that exports should be super-competitive right now and, critically, that oil prices are likely to stay down.

Even the triple zigzag (!!!) could make sense if the market were to decide to test the world's Great New Hope by sending another tsunami -- perhaps triggered by the bankruptcy of a large household name manufacturing company in early January -- to greet Mr Obama as he takes office as President.