Input Welcome RE: Growth vs Value Strategy
In What Makes the Market Go Zoom, Growth or Value? I showed that the relative-performance of growth versus value stocks on any given day have a huge influence on the day’s broad market return.
I think that this could be a very, very powerful concept, but to make use of it, we’d need to be able to make a prediction about how growth and value stocks will perform relative to the other the following day. In other words, we’d need to be able to successfully “trade” the following growth-to-value ratio:
I’ve sliced and diced the data and come up with nothing… seems like a pretty random walk to me, so I’m turning over the data to our reader community for ideas.
Whatever “strategy” was used would have to be very simple and short-term in nature (because the data set is too small to validate a strategy with complex rules or long hold-times) and trade both long and short (because there isn’t a long/short bias to the data set).
Looking forward to your input. Note: if you don’t want your results published on the blog, that’s a-ok.
Happy Trading,
ms
To stay up to date with what’s happening at the MarketSci Blog, we recommend subscribing to our RSS Feed or Email Feed.
Filed under: Random Stuff | 8 Comments
Dear Sir: sorry to say, but I have been down that path several times before, always coming back empty handed.
I admit that for the most part I used the Russell definition of growth and value. This definition is very rudimentary and very likely not the proper way to categorize stocks.
A few years ago, Nelson Freeburg, from Formula Research, published some work based on relative performance of both indexes ( the “2000″ in this particular case) and he presented impresive results based on a combo of price momentum at different time horizons. I more or less reproduced the results, but found them to be extremely erratic, with a very large number of whipsaws.
I cannot provide copy or description of the method, since both are the intellectual property of Mr Freeburg.
eber
RE to eber: thanks for the comment…so far I agree re: not being a predictable series, but I hope that someone proves me very wrong…michael
If you have some means of predicting that outcome, you are probably better off applying your method directly to your trading rather than predicting whether growth or value will do better and using that to make your trading decisions.
It seems to me that the way to use your recent result is to keep track of whether growth or value outperformance correlates to up days and use that as an intermediate-term indicator (probably more as a switch in your trading logic than a direct buy/sell signal).
RE to jkw: I see your point, but here’s why I disagree. It’s very hard to find short-term indicators (that work) that don’t rely on the normal ho hum price data (OHLCV). If we could make this work, this would be a very short-term indicator (next-day only) that is approaching the market from a radically different perspective. I’ve talked about confidence-based strategies a lot (http://marketsci.wordpress.com/2009/04/01/transactional-vs-confidence-based-trading-strategies/) and adding elements like this is bread-n-butter. michael
Hi Michael,
You may be looking at the wrong ratio. Your graph shows you whether the _level_ of the growth index is higher than the _level_ of the value index. But you’re interested in which is _outperforming_ the other at a given point in time. Perhaps you should look at the difference of that ratio, or the difference in daily returns.
RE to Joshua: funny…you gave me one of those “oh shit” moments when I first read your post…had to rethink my assumptions. Yes, the % change in the ratio will capture that (difference in % daily returns) regardless of the actual closing level of each index. The math is a little off in terms of what a change in the % difference equates to on the ratio, but close enough for research purposes. michael
Dr. Victor Canto, is his book “Understanding Asset Allocation: An Intuitive Approach to Maximizing Your Portfolio” points to more macro issue that affect size and value v. growth bias, such as inflation, taxes and regulation. Not really a ST trade ideas, but very interesting background that may provide guidance in the ST ideas. He manages money based on the philosophy with quarterly changes, not sure about the short side. Info can be found at lajollaeconomics.com.
RE to Steve – thanks for the comment – you’re right…way too long-term (and way too fundamentals-oriented) for what I do. michael