Subscribe to DSC Newsletter

The stock market has soared despite no news to support the new highs. Is it time to sell or to short? Or to buy?

Here's my advice, both as a data scientist, and as a former day trader who survived 15 years of stock market volatility  (and increased model-resistant price oscillations), mostly by not trading anymore since 2000, moving my money around from one bubble to another, and leaving with profits before each one pops. In a nutshell, being able to time these bubbles, and being a bubble chaser.

Regarding the stock market, it had its great pop in 2000 (with aftershock in 2008) and it will take many years before it becomes again an attractive (albeit dangerous), massive bubble. Maybe this will never happen again: human being have stopped building pyramids, they've stopped building cathedrals, they might one day stop flying (unless we find new sources of energy to power planes when we run out of fuel, and assuming flying is not replaced by video-conferencing and telecommuting). Maybe one day human beings will stop using stock exchanges like Nasdaq and new funding sources will be found, instead.

Anyway, I think we are now in bubble or pre-bubble stage. We could soon see permanent trading below the year 2000 all time high (inflation adjusted), like what happened in Japan after the 1990 implosion. I think DOW could easily fall below 10,000 in a matter of months or years. But I think it could as easily soar above 20,000 before experiencing another collapse (I tend to think that no two occurrences of a same pattern are identical, and thus Japan / US will see different outcomes).

The reason I think the stock market could still go up a lot, is because trading is still driven by emotions, even the Princeton data scientists who design HFT (high frequency trading) strategies put emotion into their algorithms: they translate emotions (of course, back tested and cross-validated with walk forward real tests) into C++ code. The trading rules that they come up with are certainly driven, to a large extent, by understanding human emotions and reactions to events.

So, because of this high unpredictability (will it go up or down - I don't know), my advice is to not enter the stock market now. It is tempting to short, but don't do it. I would not buy either.

Unless you can beat large professional traders, retail investors should put little to no money in the stock market. How do you grow your money then? Good question, and there is no easy answer. I grow my money by re-investing in my own growing company, but that's not an option for most of us. Some buy gold (over-prized, you can't eat gold, I'd rather buy bottled water instead), and maybe some put money in the stock market. All markets are intertwined now, so even growing economies such as China is not safe heaven for your money. The ones that are (Switzerland) are already saturated. You can still play in the stock market if you have good insider information - possibly an advantage over professional traders who can be served with jail time if caught in an insider trading scheme. Maybe one day an offshore start-up company will sell insider secrets to potential investors - I think there's a market for this type of services.

Or maybe the stock market is going up because inflation is picking up steam. So it's not actually going up, it's us going down.  If the stock market is growing fast, but not as fast as food prices, gas prices, taxes, healthcare prices, college education prices etc. then the stock market is not over-performing: it's just that we are experiencing strong inflation, inflation that the official (out-dated) indices fail to notice. But if that's the case (I don't think it is), investing in the stock market could make sense.

Also note that the DOW is at about the same level it was 12 years ago. So, inflation-adjusted, we are still much below the 2000 peak.

Under the current conditions, one stock market strategy of interest is to do a straddle: being both long and short on a same position. What I've done in the past is a bit more complicated: having multiple long and short positions (in parallel) on a same volatile stock, with various buy/short and sell/buy back dates. It also provides tax timing strategies: if you buy 1,000 shares of GE in 2012, another 1,000 in January 2013, and sell 400 later in 2013, you can write on your tax form that these 400 shares represent 100 shares from 2012 + 300 shares from 2013, or 250 shares from 2012 + 150 from 2013, whatever serves you best when you factor in tax issues.

Related articles:

Views: 3710


You need to be a member of Data Science Central to add comments!

Join Data Science Central

Comment by Vincent Granville on March 14, 2013 at 6:39am

Rule of thumb with bubbles: you can expect to consistently make 1/3 of the return to be made from  bottom to top. For instance, with the current stock market bubble, the bottom was around 7,000 a few years back. If the top is at 21,000 (say in 2013), the potential return is 200%. But the experienced bubble hunter will usually get only a 67% return in this example. Maybe he will enter at 10,000 and sell at 16,700: either before it peaks at 21,000, or on its way down after it has peaked at 21,000.

Comment by Michael Walker on March 13, 2013 at 8:12pm

Nice analysis! The evidence suggests investing in an index fund is the best strategy for non-professional investors. Most pro traders and investment managers fail to beat the market - although there are some stars who appear to do well consistently. Timing the market is very difficult. By the way, didn't the Princeton/Newport hedge fund blow-up using strategies you list? Didn't Long Term Cap Management hedge fund blow-up using purported sophisticated financial quant strategies?

The recent stock market surge may have more to do with fed policy than herd behavior.

Follow Us


  • Add Videos
  • View All


© 2018   Data Science Central™   Powered by

Badges  |  Report an Issue  |  Privacy Policy  |  Terms of Service