小股研心

suggested by Jim Cramer

-1- Market Emotions:  When emotion of bear is all over the head line of "New York Times"
-2- The Investors Intelligence Survey:  When most fund managers (60%+) tend to be bearish.  The report is posted on IBD every Thursday.
-3-Funds Cash Flow Pull Away: When funds' cash flow get in and out for several month.  Possible posted from AMG Data Service.
-4- VIX:  When VIX > 40 for three weeks.
-5- Helene Meisler's oscillator:  When the index is < -7.
-6- New Low / New High:

Tip: When buy in near market bottom, buy some stocks go up at the morning. If the market turns down, those should be less damaged.

希望股市不要再跌ㄌ

2010-03-03

Tortoise Worldview Market Model based on ETFs (as of Feb, 13, 2009)

Idea:
When news announced, market usually moves in one direction for an hour, a day, or even a week.  If we can catch the begining of the trend and act according to long or short to the same direction, then we shall have sizeable profit in relative short period of time.

Question:
How to catch this trend triggered by news?

Approach:
Assuming the market's move was triggered by news, then large amount of trade should be expected.  Thus, volume will be much larger than previous hour, day, and week on average.  It becomes an unusual volume increasing at the time period along with price change.  For example 02/12/09.

Here is a trading idea from book "Traders' Almanac" -- the first day of month.  The idea suggests that the first day of every month gives positive returns.  Reporduced results are plotted as total return of 12 months vs year (below).  It seems true to have some profits before fees on all three major indexes.  We also observe that negative returns are from bear markets and positive from bull market.

But why the first day of month is so special, not the second or third?  The following table summarizes the average return for taking the n-th day of month.  Interestingly not only the first day of month gives more than average return but also the second, the third, and the last day of month.   One supproting reason for this could be fund managers' unloading and reloading their portfolio stocks.  If you buy at Close of the previou 4-th day of month and sell at Close of the 3-rd day of month, you could have about 7% return per year on average (Dj = 6.87%, S&P500 = 7.28%, and Naqdaq = 7.37%).  This performance is outstanding because it beats DJ average annual return and close to S&P500's (DJ = 6.46%, S&P500 = 8.38% and Naqdaq = 10.96%).

It's worth noting that this strategy didn't avoid loss from bear markets.

    Day.-10  Day.-9 Day.-8 Day.-7 Day.-6 Day.-5 Day.-4 Day.-3 Day.-2  Day.-1
DJ     0.16   -0.41  -0.31  -0.40   0.17  -0.50  -0.26   0.20   0.72    1.18
SP     0.00   -0.39  -0.34  -0.10  -0.23  -0.81   0.73   0.65   0.55    1.58
NQ    -0.70   -0.56  -0.01  -0.74  -0.14  -0.09   0.59   0.22   1.00    2.59

      Day.1   Day.2  Day.3  Day.4  Day.5  Day.6  Day.7  Day.8  Day.9  Day.10
DJ     1.74    1.94   1.35   0.26  -0.31  -0.13   0.11   0.17  -0.16    0.23
SP     1.70    1.76   1.37   0.33  -0.30  -0.29  -0.05   0.23   0.46    0.17
NQ     1.28    1.83   1.43   1.03  -0.29   0.04  -0.29   0.51   0.95    0.40

      Day.11 Day.12 Day.13 Day.14 Day.15
DJ      0.94   0.02  -0.26  -0.35  -0.17
SP      0.74   0.32  -0.33  -0.13   0.39
NQ      0.36   0.33  -0.26  -0.39   0.74

Here is a daytrade pattern for Open.  In general, buy at Open is NOT a good idea -- most of trades end up losses.  However, there are few opportunities when Open is very low or high. 

Take SPY as an example, buy at Open when it opens -2% or more and exit at some time points before Close.  Almost every trades have much higher High to allow you to exit for "little" profits, and the Low are not far from Open which says the loss can be limited.  In order to have sizeable profit, conside to laverage the position, or trade on Ultra-ETFs, or individual stocks. 

Another example is to buy at Open when it opens > 2%.  This will also be good trades because it simply trades on momentum.

Following plots illustrate the two examples above.  Click plot to enlarge.

Market Chart of Post-Election Years

2009 is a Post-Election year.  Should the market go up or down?  

From past 20 Post-Election years, 10 were up and 10 down.  However, 7 out of 10 down year are of Republican President.  There is better chance to bet 2009 UP because Obama is Democrat. (see large plot)

Hypothesis: Hot January industries beat S&P in next 11 month

Result:
This study uses S&P500 stocks which are categorized into 59 industries.  Industry performance is calculated as weighted average of percent return with weights (stock market cap) / (industry market cap).  For each year, Industry performance on January against rest 11 month performance is plotted, and top 5 industries of January are dots colored in red, and we expect those red dots to have higher returns on rest 11 months.

Conclusion:
In Bull years, top industry of January tend to have better performance than others.  However, it is not better and can be worse in Bear year.  This hypothesis is true in condition of Bull years.

Plots: