Thursday, October 8, 2009

Is The Model "T" Forecast Too Low?

The response to the blog has been very positive. But several people have questioned whether the 580 number (S &P 500 index) is too low when the index is currently above 1050 (see the first post). I have to agree, I think it might be too low. But the great thing about the Model “T” is that the model doesn’t think, it doesn’t read the newspaper and it doesn’t surf the Internet. It just forecasts based on inputs, with no emotional bias.

If 580 is indeed too low, what factors could cause this? First, the model does not take into account government intervention. And we have seen extensive government intervention in the banking system, stimulus bill, General Motors takeover and programs to boost the auto and housing industry. It would be difficult to add these factors to the model because they have never been tried before and their long-term impact is uncertain.

Second, all models attempt to predict the future based on how things have operated in the past. The Model “T” is based on factors in the transportation market. The transportation market may not be as significant to the economy as it has in the past due to the decline in domestic manufacturing. Therefore a large drop in the transportation segment factors will not have the same negative impact to the model as previous cycles.

If the bottom of this cycle was the 676 low reached in March, then maybe the Model “T” did its job and was off by only 96 points (not too shabby). Of course this assumes that the stock market and economy hit bottom then and with the help of the successful government stimulus efforts began a new recovery cycle.

But the Model “T” does have a timing aspect to it and it indicates that the stock market fell too low, too fast, in Q1. This was due to emotional panic caused by the financial crisis. Once that panic stopped, optimism has replaced it and the market has risen in the words of Nouriel Roubini (Dr. Doom) “too much, too soon, too fast”. The stock market has been driven by more emotional factors than usual over the last 12 months and understandably so.

While many of analysts are predicting continued good things for the stock market, Robert Pracher (Elliot Wave International Inc.) recently said his model is forecasting the S & P 500 will fall “substantially” below the March bottom of 676. Economists predicting a market correction cite weak consumer spending, a fragile economy and continued problems in the banking and housing industries as reasons.

If government stimulus is important and the Model “T” is based on the transportation industry, a relevant question is: How much has the government’s actions benefitting the transportation industry? The answer is, not much. FTR Associates (a respected industry forecasting firm) reports that truck freight volumes continue to drop with no significant improvements until mid-2010. Railroad freight is still down more than 10% from last year.

It appears the government’s economic interventions were successful at stabilizing the economy and the stock market, but they may have extended out the down cycle. This opinion is supported by Chris Ciovacco (greenfaucet.com) who predicts the stock market may not fully correct until well into 2010 based on his model. He also infers that 676 may not be the bottom. At this point the 580 forecast of the Model “T” does seem low; however it is interesting that the model is providing a forecast similar to other, more sophisticated, statistical models.

Disclaimer Statement

The information contained in this blog is for strictly discussion and reference purposes only. In no way and under no circumstances should the information presented here be intended as investment advice. Statement s made on this blog do not represent a recommendation on buying or selling equities or securities nor which ones to trade. Please make your own responsible investment decisions based on your own research.
The information in this blog is solely the opinion of the writer (except for comments made by people to the posts or references in the posts attributed to other people).

No comments:

Post a Comment