buy more stocks of a stock that went down

Models
1

I want to implement the following:

Start with the normal fixed position size.
After 4 Weeks, the system should buy more (100% of the original position size in $ Terms) if the stock went down.

Is that possible and if how?

Best Regards
Andreas

2

Run the original simulation without any changes. Call it “Port X”

Then run another simulation that’s exactly the same but with the following rule in formula weight rebalancing:

Eval (close(0)/close(22) < 1 and PortfolioOpenBar (“Port X”) > 19, 2, 1)

Note that this will keep doubling your position every rebalance period if the stock keeps going down.

3

wow, thank you.

I run the test and I would have sweared before the test, that this would improve the system.
It did not it made it 10% worse ann.
WOW.
there you go, intuition without a backtest is not a good lead.

Thank you!!!

Best Regards

Andreas

4

Hi Andreas, I understand you utilize momentum as a key part of your models, and you’ve probably tried this - but an alternative approach to try to “buy the dip” might be to put a short term mean reverting signal into the model and see if it achieves a similar effect by increasing the ranking of stocks that have declined in the past week or few weeks. Something like close(0)/close(n) where n is number of days, like 20, 15, 10, 5, etc. I’ve found this almost always helps backtests when implemented with strong longer term positive momentum - although it increases turnover and shuffles stock rankings more depending on the weight given.

edits: Also I should’ve added, the Bollinger Band function in p123 can be utilized to do something similar as above, taking volatility into consideration.

5

Andreas, I thought the saying is: “Let the winners run”.
Eval (close(0)/close(22) > 1 and PortfolioOpenBar (“Port X”) > 19, 2, 1)

6

Hi SpacemanJones,

thank you, tried it out, equal weighting with no further purchases still works the best, hmmmmm…

Hi Georg,

seems to be the case :wink:

Regards

Andreas

7

Or the corollary to “Let the winners run”…“don’t catch a falling knife”. I would suspect that “buying the dip” works better on the board market than individual stocks. Individual stocks are more likely to spend a longer period of time falling than the broader market as a whole. If there is naturally upward momentum in the stock then “buying the dip” may work.

8

When I back tested on another platform there did seem to be a some alpha in buying stocks that performed the worst over the past 5 years with good fundamental valuations and low volatility in the past few months. Just taking advantage of situations where stocks were excessively beaten up. I haven’t tried than on Portfolio123 though.

9

Hi Schm1347, that alligns with backtests from academics, momentum works on a shorter timeframe (up to 12 Months)… but not on longer timeframes…

Best Regards

Andreas