StockFetcher Forums · General Discussion · backtesting ideas<< 1 2 3 4 >>Post Follow-up
tonyandrosa1975@gmail.com
1 posts
msg #156926
Ignore tonyandrosa1975@gmail.com
6/5/2021 7:51:50 AM

can someone tell me me where i can backtest indicators after they move from one level to another, and the profit or loss after a certain number of days after such occurrance.

Example: profit or loss 30 days after MACD has a bullish crossover.
i like to backtest it back to 10 years and find out how many times it was profitable
thank you for your help

novacane32000
331 posts
msg #156956
Ignore novacane32000
modified
6/9/2021 6:26:32 PM

I use Trendspider. It is good for backtesting simple filters as you mention.

Here are the results of the MACD(12,26,9) signal line crossing above 0 as an entry and exit would be the signal line crossing below 0.

Daily for SPY

Over the past 4 years there were 11 positions

6 winners (avr win 9.38%)
5 losers (avr loss -2.48%)
Net profit 49%
Max drawdown -14.3%

So nice results for the investor type trader especially with the rather pain free low drawdown but the asset performance for SPY during the same period was 77% compared to the 49% gain from the strategy.


So you traded lower performance for a nice equity curve.

If you tell me exactly how you want to trade it, I can see if Trendspider can do the testing.


Cheese
1,374 posts
msg #157001
Ignore Cheese
6/15/2021 6:13:54 PM

@novacane32000

I didn't want to interrupt your original conversation, but this pause may be a good seque way.

Are there any backtest results that you've already done and that you are willing to share?

For example, did you find that setting MACD and other indicators with longer parameters
may be better trading or money management practice, e.g. MACD longer than 12,26,9?

Thank you.


redversa721
157 posts
msg #157003
Ignore redversa721
modified
6/15/2021 9:39:42 PM

I tried the following
My Watch Lists: REG_1
Reg1, contains the top 5 stocks from each sector in finviz
MACD(12,26,9) Crossed Above 0
UO(7,14,28) Is Above 30. [ultimate oscillator]

Initial Cap : 30K
SL : 16%
Close Position after : 145 days
Max Portfolio size : 3

Time Period : 1/1/2012 - 12/31/2021


Results :
Total Profit: $472,413
Total Trades: 53
Capital Growth: 1,575%
Profit Trades, % of Total: 70%
Profit Factor: [?] 6.82
Avg Trade Duration, days: 122
Payoff Ratio: [?] 2.95
Avg Profit per Trade: 20.8%
Max Drawdown: 28%
Avg Profit per Day: 0.17%
Max Drawdown, $: 20,567
Avg Market Impact: [?] : Minor

Restoration Factor: [?] 22.97
Overall Viability Score: [?] : 10 / 10
Avg Annual Return: 38.9%

Cheese
1,374 posts
msg #157004
Ignore Cheese
6/15/2021 10:13:13 PM

https://www.stockfetcher.com/forums/General-Discussion/backtesting-ideas/156926
redversa721 modified 6/15/2021 9:39:42 PM

top 5 stocks from each sector in finviz
MACD(12,26,9) Crossed Above 0
UO(7,14,28) Is Above 3
=========================================================================

Thank you, red !

redversa721
157 posts
msg #157005
Ignore redversa721
6/15/2021 10:48:01 PM

Np. Thanks to all your help. in the forums..
if you need me to create any more backtests, lmk and I can run them.

I use marketinout for backtesting.

Cheese
1,374 posts
msg #157006
Ignore Cheese
6/15/2021 11:28:31 PM

red,

Thanks for your kind offer.

I don't have any backtest requests. I only have a couple of simple questions.
Just answer off the top of your head is good enough.

1. With MACD, do you prefer Fast Line crossed above Slow Line (i.e. even when MACD < 0}
or do you insist on MACD > 0

2. In general in your experience do you prefer MACD longer than 12,26,9 or shorter?

If you feel like laughing, go ahead. I don't mind. I ask people silly questions all the time.

redversa721
157 posts
msg #157007
Ignore redversa721
6/16/2021 12:13:35 AM

Hi Cheese

I extensively backtest random combinations to see which one has a high chance of success. After getting burnt trying to get rich fast, I have moved on to a more sedentary investing strategy.

I have not used the previously mentioned strategy, just back tested it on account of the request.

Bread and butter is I use a stoch and a super trend based strategy on just a select set of good quality stocks.

for some options based strategy, I use CMF and a MA based strategy. I do not get a lot of values but so far, the few that I get have been working out ok. [Still testing it on live account after paper trading for six months]

But to answer your question honestly, I actually do not know the details or the differences between the various stoch combinations and that's why I back test them a lot since my brain kinda fades looking at charts.

nibor100
1,031 posts
msg #157008
Ignore nibor100
6/16/2021 5:52:26 AM

@redversa721,

Regarding the backtest stats you posted:

a. 3 trades at a fixed $10K per trade, or 3 trades max at 33% of available equity, or $30K per trade?

b. Is your group of stocks from the Finviz sectors only based on the most recent 5 top stocks per Finviz sector or is the backtest choosing from an updated historical group at the onset of each new trade?

c. What made you choose a Stop Loss of 16% and a max trade duration of 145 days? I'm not sure I would ever choose those 2 parameters for any initial backtest.

Thanks,
Ed S.

redversa721
157 posts
msg #157009
Ignore redversa721
modified
6/16/2021 9:54:42 AM

Hi Nibor

It is a total of 30K , so begins with 10K per stock and you reinvest gains, so end of year 1, let's say 10% gains. Then per stock will be 11K starting end of year 1.

so to answer your question, 3 max equities at 33% max portfolio

b. That is an excellent question, because I have not been able to auto update my watch list. so make life simple, I have only picked stocks that are more than i believe 10B as of today. so yes, to highlight your point, those stocks may not have been chosen in the previous year since it would not have been > 10B cap 5 or 6 years ago. To minimize this effort of updating my watch list each year, I went the easy way forward and only picked high quality stocks from each sector, sorted by capitalization
Large cap > 10b, EPS growth over 5 year >10, Div yield > 0, NPM> 0, PE >5, EPS next 5 years> 5%, ROE > 5% , forward p/e > 5, ROI > 10, EPS curr year > 5

so great point but there is a bias in the result, which I am hoping to minimiz using long lasting good quality stocks

c. I tried multiple combinations and this gave me the best result over 10 years. I have a list of standard combinations that i run my backtest against to ensure that i am comparing multiple backtest and strategies against the same sample.

let me know if I answered the question.






StockFetcher Forums · General Discussion · backtesting ideas<< 1 2 3 4 >>Post Follow-up

*** Disclaimer *** StockFetcher.com does not endorse or suggest any of the securities which are returned in any of the searches or filters. They are provided purely for informational and research purposes. StockFetcher.com does not recommend particular securities. StockFetcher.com, Vestyl Software, L.L.C. and involved content providers shall not be liable for any errors or delays in the content, or for any actions taken based on the content.


Copyright 2022 - Vestyl Software L.L.C.Terms of Service | License | Questions or comments? Contact Us
EOD Data sources: DDFPlus & CSI Data Quotes delayed during active market hours. Delay times are at least 15 mins for NASDAQ, 20 mins for NYSE and Amex. Delayed intraday data provided by DDFPlus


This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.