I'm a holder but willing to learn a bit of quantitative trading

L

Luiz

I just started reading a bit of it, but everything is so abstract to me that I can't use it properly. So, I started to build my own logic with something I can understand, and it would be great if you could give me your advice.

The idea is mean reversion in a low beta portfolio. I'm using Excel to calculate the 100-day mean and the standard deviation. Then, I'm calculating how far the stock is from the mean and dividing it by 10% of the standard deviation. If a stock is down 10% and has a standard deviation of 2, it will give me 0.5 as an answer(10%/(0.1*2)). If a stock with a 0.25 standard deviation is 5% down from the mean, it would give me 2 as an answer, so I would buy the second one instead of the first.

I'm keeping 10% of my funds to buy more if something goes lower. If my fundamental analysis was good enough, I shouldn't see a huge crash in any stock.

I guess this is it. Any advice on risk management would be good too.
 
Luiz said:
I just started reading a bit of it, but everything is so abstract to me that I can't use it properly. So, I started to build my own logic with something I can understand, and it would be great if you could give me your advice.

The idea is mean reversion in a low beta portfolio. I'm using Excel to calculate the 100-day mean and the standard deviation. Then, I'm calculating how far the stock is from the mean and dividing it by 10% of the standard deviation. If a stock is down 10% and has a standard deviation of 2, it will give me 0.5 as an answer(10%/(0.1*2)). If a stock with a 0.25 standard deviation is 5% down from the mean, it would give me 2 as an answer, so I would buy the second one instead of the first.

I'm keeping 10% of my funds to buy more if something goes lower. If my fundamental analysis was good enough, I shouldn't see a huge crash in any stock.

I guess this is it. Any advice on risk management would be good too.
More...
Please check:
I think there is a logic error in your algorithm:
a smaller SD (for the same PnL%) gives a higher score.
Isn't that an illogical (wrong) result?
 
For quantitative trading I would recommend Quantopia. The company doesn't exist anymore but the resources do: lectures, YouTube videos, etc.

There are more sources but with that you will have enough for a while.
 
Quanto said:
Please check:
I think there is a logic error in your algorithm:
a smaller SD (for the same PnL%) gives a higher score.
Isn't that an illogical (wrong) result?
More...
the idea is if a stock oscilates between 2% and -2% of the mean and its down 5%, in my opnion its more logical it go up than a stock down 5% wich oscilates between 20% and -20%
 
DarkTemplar said:
For quantitative trading I would recommend Quantopia. The company doesn't exist anymore but the resources do: lectures, YouTube videos, etc.

There are more sources but with that you will have enough for a while.
More...
I found it, i've tried before to search something like this on youtube but everything i found wasnt good. Ty
 
Luiz said:
I'm keeping 10% of my funds to buy more if something goes lower. If my fundamental analysis was good enough, I shouldn't see a huge crash in any stock.

I guess this is it. Any advice on risk management would be good too.
More...
Where is your risk control?
You shouldn't see a huge crash in any stock but you might. In a bear market pretty well everything tanks.
You are relying on your fundamental analysis, is it better than the institutions who get caught every so often? Gotta remember GAAP allows accountants a lot of leeway.
 
The price should follow the profit; all actions in my sample space have not recorded losses in the last 10 years. Therefore, as long as nothing happens to them, the price should return to the average, assuming the market is efficient. I haven't done backtesting because I recently started learning Python and still don't know how to do it. That's why I'm using a small amount of money to see how the strategy performs going forward.
2rosy said:
what makes you think a stock price mean reverts? did you test that first?
More...
 
deaddog said:
Where is your risk control?
You shouldn't see a huge crash in any stock but you might. In a bear market pretty well everything tanks.
You are relying on your fundamental analysis, is it better than the institutions who get caught every so often? Gotta remember GAAP allows accountants a lot of leeway.
More...
In a bear market i would just keep my stocks as i would in my long-term portfolio, my risk control rn is not alocating too much capital in one stock and verifying if the company remains healthy
 
Luiz said:
In a bear market i would just keep my stocks as i would in my long-term portfolio, my risk control rn is not alocating too much capital in one stock and verifying if the company remains healthy
More...
In that case I'd suggest you put your money in SPY. Chances are you won't outperform the index.
 
You must log in or register to reply here.
Back
Top