Howard’s Journal

[Howard Lask]

Thanks Scott.

This has been intriguing me so I devised a test. Please throw rocks and challenge me on the approach and conclusions.

I designed a simple rotation system on the RUT. 20 positions selected using randomised ranking, rotated monthly. Test period 1/1/04 to 12/31/20.

I then optimised for rotation day – from 0 to 20 where 0 is the last day of the month, 1 is the first etc (some months may have more than 21 trading days but I don’t think this will make a big difference and more importantly I couldn’t figure out how to test for variable month lengths)

I repeated this 250 times for each rotation day so 21 x 250 = 5250 tests

Results are included in the attached table.

Very consistent results from one day to the next with days 7 and 9 marginally poorer and days 12 and 14 marginally better. I’m no statistician but I doubt these are statistically relevant

My conclusion – open to challenge – is that there does not appear to be an end of month entry edge

[Nick Radge]

Quote:

My conclusion – open to challenge – is that there does not appear to be an end of month entry edge

Which begs the question as to why your system test showed deterioration…

[ScottMcNab]

Seems I am in minority….been a while since I looked at it so I better go back and test again…I have developed some markedly different systems in last few months compared to the ones used over the last few years so may be quite different

[JulianCohen]

have you tried the same test on your actual system Howard? That might be the next test to do

[Howard Lask]

Julian Cohen wrote:

have you tried the same test on your actual system Howard? That might be the next test to do

Nick Radge wrote:

Which begs the question as to why your system test showed deterioration…

I had been stress testing the system by changing the rotation day and saw a large deterioration in performance. My hypothesis was that I had over-optimised so performed this test as confirmation.

These results confirmed my hypothesis and should not expect a month end rotation day to perform significantly differently from any other rotation day.

Time well spent as I have learnt something that is valuable for future system development

[JulianCohen]

Ah got you.

bin that one then…on to the next one ha ha

[Howard Lask]

I have paused research into the R1K system as I’m still not able to get robust results and have turned my attention to MR systems, initially by converting the approach for my MOC systems.

June MOC in the R1K Long system took a hit earlier in the month. The API somehow got disconnected after orders were placed and corresponding MOC orders were not entered. I only realised after the close, manually placed orders to exit at the open and took some losses due to overnight gaps. Another item to add to the daily checklist!

June 2021

MOC Systems
R1K Long: -8.30%
R2K Long: -0.37%
R1K Short: +0.64%
R2K Short: -0.63%

Long Term Systems
Nasdaq Momentum: +7.10%
R2K DTT: +3.13%

[Howard Lask]

July 2021

MOC Systems
R1K Long: +1.97%
R2K Long: -9.52%
R1K Short: -3.52%
R2K Short: -4.39%

Long Term Systems
Nasdaq Momentum: +15.88%
R2K DTT: -3.99%

Total Account: +2.00%

[Howard Lask]

August 2021

MOC Systems
R1K Long: +1.44%
R2K Long: -0.78%
R1K Short: -5.22%
R2K Short: +11.67%

Long Term Systems
Nasdaq Momentum: +2.98%
R2K DTT: +6.44%

Total Account: +3.28%

[Howard Lask]

September 2021

MOC Systems
R1K Long: -1.81%
R2K Long: +1.75%
R1K Short: +1.30%
R2K Short: -0.46%

Long Term Systems
Nasdaq Momentum: -2.39%
R2K DTT: +0.36%

Total US Account: -3.54%

ASX 100 Rotation: -1.56%

I have been working on longer term systems over the past few months and have gone live with an ASX Rotation system in September. I’m also currently overhauling my DTT system and will be moving to a weekly rather than daily timeframe. Performance is comparable and will save me a few minutes a day.

[Howard Lask]

October 2021

US Systems
MOC Systems
R1K Long: +0.70%
R2K Long: +0.60%
R1K Short: -1.73%
R2K Short: +1.86%

Long Term Systems
Nasdaq Momentum: +2.73%
R2K DTT: +3.87%
WTT: +3.99%

Total US Account: +2.29%

ASX 100 Rotation: +1.91%

[Howard Lask]

November 2021

US Systems
MOC Systems
R1K Long: +1.47%
R2K Long: +4.01%
R1K Short: +5.90%
R2K Short: -0.94%

Long Term Systems
Nasdaq Momentum: -12.17%
R2K DTT: -7.41%
WTT: -10.24%

Total US Account: -5.66%

ASX 100 Rotation: +7.12%

Taken to the cleaners this month in my US Momo systems – heavy exposure to MRNA caused most of the damage

[RobertMontgomery]

Great job on the MOC short system return, I have to find a setup that can rival my long systems.

[Howard Lask]

December 2021

US Systems
MOC Systems
R1K Long: -0.92%
R2K Long: +4.46%
R1K Short: -2.47%
R2K Short: +1.97%

Long Term Systems
Nasdaq Momentum: -9.41%
R2K DTT: +2.11%
WTT: -3.55%

Total US Account: -2.67%

AUS Systems
ASX 100 Rotation: +16.16%

Full Year 2021

US Systems
MOC Systems
R1K Long: +30.81%
R2K Long: +60.97%
R1K Short: +8.63%
R2K Short: +31.13%

Long Term Systems
Nasdaq Momentum: +0.82%
R2K DTT: +22.11%
WTT: -14.21%

AUS Systems
ASX 100 Rotation: +24.83%

[Nick Radge]

Nice work Howard!!

[Author]

Posts