Tuesday, June 26, 2007

Should I Chase Stocks After They've Risen?

It's a dilemma. You have a great trade idea, but before you can get on board, the stock or index makes its move. Now you've got to decide to wait for a pullback or chase the rising prices.

I'm finding the money flow data to be valuable in addressing this question, particularly when moves are to the upside.

Recall that I track 40 highly-weighted S&P stocks that are evenly divided among eight sectors: Materials, Industrials, Consumer Discretionary, Consumer Staples, Healthcare, Energy, Technology, and Financial. Each day I assess the raw money flows into these 40 issues.

As a quick reminder, money flow starts with each trade in each stock. The price of the stock is multiplied by the volume of that particular trade. If the trade occurred on an uptick, the dollar volume is added to a cumulative total. If the trade occurred on a downtick, the dollar volume is subtracted from the total. At the end of the day, we look at the cumulative total for each stock and sum across the 40 stocks.

Going back to 2005 (N = 617), when the average money flow over a five day period is below 75 million (N = 121), the next five days in the S&P 500 Index (SPY) average a gain of .37% (73 up, 48 down). When the average money flow for the 40 stocks exceeds 150 million (N = 180), the next five days in SPY also average a gain of .37% (117 up, 63 down).

By contrast, all other occasions in the sample--occasions in which average five-day flows are between 75 million and 150 million--average a five-day gain of only .06% (174 up, 142 down). In other words, when we look at a five-day horizon, we see bullish implications when money flows for the S&P stocks are very weak (reversal effect), but also when they're very strong (momentum effect). Gains are subnormal following periods that don't feature either very weak or strong money flows.

Another way of looking at the data is to examine rising five-day periods in SPY as a function of money flows. Going back to 2005, we've had 268 periods in which SPY has gained more than .50% over a five-day period. Following these periods, the average five-day change in SPY is -.08% (136 up, 132 down).

When we divide the sample in half based on money flows, however, a pattern emerges. Five-day gains on strong money flows (N = 134) average a subsequent five-day gain of .13% in SPY (77 up, 57 down). Five-day gains on weak money flows (N = 134) average a five-day loss of -.29% in SPY (59 up, 75 down).

What that suggests is that traders might not want to chase rises on weak money flows, but rises on strong flows tend to persist in the short run. Note in my latest post to the Trading Psychology Weblog that we've had very strong money flows in the 40 S&P stocks and that this corresponded to a persistent price rise in the index. In a rising market, it has paid traders to go with the money flows.

RELATED POSTS:

Dow Returns Following Extreme Money Flow Days

What's Behind the Bull Market

4 comments:

m said...

As is typical, interesting and useful information. Have you (or do you where I can find) done the same analysis over longer time frames- multiple days, weeks, months? Does the same patterns hold?

Brett Steenbarger, Ph.D. said...

Hi M.,

I've carried out the analysis out to 50 day holding periods and found value in the money flow data. My general conclusion is that money flow behaves somewhat similarly to the NYSE TICK: it's relative shifts in levels and how those impact price that is key.

Brett

nichols_ryan said...

Having you analyzed putting bollinger bands to the MF, to see if low probability flows are more directly related to market returns?

Brett Steenbarger, Ph.D. said...

Hi Nichols,

Excellent idea; I'll need to look into that. Thanks!

Brett