Discuss Market Comments 10/27

I like observing the seasonality charts. They are influential in my decision making at times. And you really like to put them into your commentary as if they are quite useful to your market timing.

However, looking back over September and October (so far) of this year, it seems like they were poor predictors. And how the markets move is really dependent on the current market conditions.

I wonder if there's a way to look at the last 6-12 months of seasonality data, comparing the actual monthly performance to that in the many years of seasonality data, and come up with a rating for the recent "predictiveness" of the longer-term seasonality statistics. Do you supposes that there's a correlation coefficient that moves with time that would help us weigh the recent and on-going predictiveness of the seasonality data?
 
I like what Jason from sentimenTrader.com has said about seasonality. He says it is not a primary indicator. It is more like a breeze while riding a bike. If the breeze is in your face, it will be more difficult to move forward, but it won't stop you. If the breeze is at your back, it makes going forward much easier.

That means if you do you analysis and you think the market is bullish, a positive seasonality will give you the extra confidence in your decision. If the seasonality is negative it may take a little more to push the market. But it should never be used as a sole indicator.

If a day or week is positive 65% of the time, that is strong positive seasonality, but that also means it is negative 35% of the time - no sure thing.

The only time I might use seasonality as a primary decision maker is around holidays; particularly the Friday after Thansgiving, and the week between Christmas and New Years. When you start seeing 70% positive or negative, it is tough to want to fight it.
 
Back
Top