Seasonal Calculator Launch
Moderator: Aitrus
-
- Posts: 62
- Joined: Mon Jun 05, 2017 6:38 pm
Re: Seasonal Calculator Launch
Put 5037 after "ID=" here in the address:
http://tspcalc.com/seasonal.php?ID=5037&years=04-17
http://tspcalc.com/seasonal.php?ID=5037&years=04-17
Re: Seasonal Calculator Launch
There probably won't be any new developments for a while. I dusted off my old Nintendo 64 and my evenings have been sucked into Harvest Moon and Super Mario. So if you need me on a Tuesday evening, I am probably planting virtual turnips and wooing a virtual lass. Either that or my wife and I are touring one of the asian countries around here and engorging in local cuisine, culture, and alcohol.
Just be glad my super nintendo is broken. If that was working, you might never hear from me again!
I think the next tspcalc update will be 2.0 (a total overhaul), and like 6 months from now.
Just be glad my super nintendo is broken. If that was working, you might never hear from me again!
I think the next tspcalc update will be 2.0 (a total overhaul), and like 6 months from now.
Owner/creator of TSPcalc.com - "Know your numbers"
Re: Seasonal Calculator Launch
Don't forget Donkey Kong 64.
Re: Seasonal Calculator Launch
nrialto wrote:Don't forget Donkey Kong 64.
Never played that one. Back in college, we had a 400 seat auditorium. We went in there on a Saturday once and hooked up Mario Kart 64 to the theatre sized projector. That was fun.
Owner/creator of TSPcalc.com - "Know your numbers"
- Longstreet
- Posts: 44
- Joined: Mon Jun 17, 2013 10:39 am
Re: Seasonal Calculator Launch
Goldeneye and Perfect Dark FTW. Must have been an awesome projector setup.
Re: Seasonal Calculator Launch
So a question. If you look at the current leader #6161 the calcultor shows his mtd return at .52%.
But if you use tspcenters return charts it should only be .48%.
G fund: 06/01 - 6/12= .07%
F fund: 06/13 - 6/21= .41%
Is the calculator month to date still off a little?
But if you use tspcenters return charts it should only be .48%.
G fund: 06/01 - 6/12= .07%
F fund: 06/13 - 6/21= .41%
Is the calculator month to date still off a little?
Those who 'abjure' violence can do so only because others are committing violence on their behalf.
Re: Seasonal Calculator Launch
Three one hundredths of a percent is well within the margins for normal tracking error, I believe.
Seasonal Musings 2022: viewtopic.php?f=14&t=19005
Recommended Reading: http://tspcenter.com/forums/viewtopic.php?f=14&t=13474
"It's not what happens to you, but how you react to it that matters" Epictetus
Recommended Reading: http://tspcenter.com/forums/viewtopic.php?f=14&t=13474
"It's not what happens to you, but how you react to it that matters" Epictetus
Re: Seasonal Calculator Launch
mmmmmbeer wrote:So a question. If you look at the current leader #6161 the calcultor shows his mtd return at .52%.
But if you use tspcenters return charts it should only be .48%.
G fund: 06/01 - 6/12= .07%
F fund: 06/13 - 6/21= .41%
Is the calculator month to date still off a little?
Three things:
1. Mmmmbeer, that is not how you calculate compound returns. Try this:
Return for the month = (1.07 * 1.41) - 1
2. You are skipping a day. One time period should end on the 12th, and the other should begin on the 12th. As you have it, you skipped a day worth of gains.
3. The TSP publishes 4 decimal points. To get the exact calculation, you have to use all 4. Then round it down to 2 for easier reading. That is how tspcalc does it, and I imagine that is how this website does as well.
Owner/creator of TSPcalc.com - "Know your numbers"
Re: Seasonal Calculator Launch
mjedlin66 wrote:mmmmmbeer wrote:So a question. If you look at the current leader #6161 the calcultor shows his mtd return at .52%.
But if you use tspcenters return charts it should only be .48%.
G fund: 06/01 - 6/12= .07%
F fund: 06/13 - 6/21= .41%
Is the calculator month to date still off a little?
Three things:
1. Mmmmbeer, that is not how you calculate compound returns. Try this:
Return for the month = (1.07 * 1.41) - 1
2. You are skipping a day. One time period should end on the 12th, and the other should begin on the 12th. As you have it, you skipped a day worth of gains.
3. The TSP publishes 4 decimal points. To get the exact calculation, you have to use all 4. Then round it down to 2 for easier reading. That is how tspcalc does it, and I imagine that is how this website does as well.
Ahhh... cool. Thanks for the clarification bro. I'm learning all kinda stuffs!
Those who 'abjure' violence can do so only because others are committing violence on their behalf.
Re: Seasonal Calculator Launch
Heads up, I found an error in the trading calendars. This error affected any "last trading day" trade for months less than 31 days long. If you had saved a copy of the tspcalc trading calendar, please go to the website again to get the corrected calendar.
Thanks,
-Matt
Thanks,
-Matt
Owner/creator of TSPcalc.com - "Know your numbers"
Re: Seasonal Calculator Launch
I'm curious to how people are picking their horse in this race. Are people choosing by total compound returns, the yearly average, time in stocks? Are people including all years?
There was talk about how 2009 was such an odd year, and the top strategies have returns for that year of over 100%, is it worth keeping that in mind when choosing the best strategy?
Personally I took '09 out and resorted the top 100 strategies, but I'd love to know how other people are going about it.
There was talk about how 2009 was such an odd year, and the top strategies have returns for that year of over 100%, is it worth keeping that in mind when choosing the best strategy?
Personally I took '09 out and resorted the top 100 strategies, but I'd love to know how other people are going about it.
Your Free Financial Advisor. Zero trading costs
https://www.wisebanyan.com/referral/92F0PFFWI
https://www.wisebanyan.com/referral/92F0PFFWI
Re: Seasonal Calculator Launch
I'm taking the lazy man's approach and going with the number one on the list. Luckily, anytime a new strategy tops the list, it's usually only one or two different IFT's. It's especially easy to change strategies thanks to mjedlin66 adding a trading day calendar.
-
- Posts: 79
- Joined: Sat Apr 17, 2010 7:13 pm
Re: Seasonal Calculator Launch
I know I advocated for ranking by average earlier but I am having seriously having second thoughts about this. I knew that Average would amplify outliers but I didn't think it would make this big of a difference. What this is doing is advocating for a couple extreme values in each strategy even though it may diminish the quality of the rest of the return. Maybe using a median would be a better metric? Or maybe an average or median after removing statistical outliers? I'm not sure, does anyone have any thoughts?
I brushed up on some statistics and I think Aitrus has the right idea with the CAGR that he uses in his seasonal thread. This is the same metric that the tsp website uses to track returns because it says more about the performance. This would definitely be better than ranking by a straight up Compound return because it would compensate for compounding bias and offer a "smoothed" value. The downside is that this would have the same as averaging in that it would hide giant swings that a strategy might have. Which isnt really what you want in a seasonal strategy. But I still believe this would be better than the current compound percentage.
To compensate for this outlier amplification effect, I think measuring the standard deviation of each strategy would be a good way to measure the size of swings. This would give us a sense of volatility and risk in each strategy.
One other thing after figuring out a good performance metric is a reliability metric. I'm a little hazy on this but maybe some sort of a r-value, chi square value, or maybe a t-test?
In summary: It would be nice to having metrics to measure performance, risk/variability, and reliability. Those three numbers would give us a very good sense of strength of each strategy. The hard part is figuring out how to measure each. Does anyone have any thoughts on any of this?
mjedlin, im not trying to hijack your thread or even tell you what to do next. Im just trying to start a conversation on measuring. Really appreciate the work you have done, and im curious to know what you think of the above!
I brushed up on some statistics and I think Aitrus has the right idea with the CAGR that he uses in his seasonal thread. This is the same metric that the tsp website uses to track returns because it says more about the performance. This would definitely be better than ranking by a straight up Compound return because it would compensate for compounding bias and offer a "smoothed" value. The downside is that this would have the same as averaging in that it would hide giant swings that a strategy might have. Which isnt really what you want in a seasonal strategy. But I still believe this would be better than the current compound percentage.
To compensate for this outlier amplification effect, I think measuring the standard deviation of each strategy would be a good way to measure the size of swings. This would give us a sense of volatility and risk in each strategy.
One other thing after figuring out a good performance metric is a reliability metric. I'm a little hazy on this but maybe some sort of a r-value, chi square value, or maybe a t-test?
In summary: It would be nice to having metrics to measure performance, risk/variability, and reliability. Those three numbers would give us a very good sense of strength of each strategy. The hard part is figuring out how to measure each. Does anyone have any thoughts on any of this?
mjedlin, im not trying to hijack your thread or even tell you what to do next. Im just trying to start a conversation on measuring. Really appreciate the work you have done, and im curious to know what you think of the above!
Re: Seasonal Calculator Launch
Even a small tweak of being able to eliminate a year, or manually pick which years you want returns for in order to weed outlier years out of the equation would help a lot to eliminate bias towards strategies that have such a high compound returns or average caused by a volatile year.
If I look at the top 100 strategies for 2004-2008, none of those calculations is also in the top 100 for 2010-2017.
If I look at the top 100 strategies for 2004-2008, none of those calculations is also in the top 100 for 2010-2017.
Your Free Financial Advisor. Zero trading costs
https://www.wisebanyan.com/referral/92F0PFFWI
https://www.wisebanyan.com/referral/92F0PFFWI
Re: Seasonal Calculator Launch
Understanding that some years are outliers, if you look at the top strategy (6769), it's worst year netted 12.58 %... Not bad.
But if you look at the top strategy from 2010-current (#6003), it's worst year was -25.63%
One thing I like about the top strategy is that it did so well in 2008, meaning that there's a good chance that particular strategy does well in being out of the market during the worst parts of the year (I think?)
But if you look at the top strategy from 2010-current (#6003), it's worst year was -25.63%
One thing I like about the top strategy is that it did so well in 2008, meaning that there's a good chance that particular strategy does well in being out of the market during the worst parts of the year (I think?)
Fund Prices2024-03-28
Fund | Price | Day | YTD |
G | $18.15 | 0.05% | 1.05% |
F | $19.08 | -0.06% | -0.74% |
C | $82.21 | 0.11% | 10.55% |
S | $82.43 | 0.30% | 6.92% |
I | $42.57 | -0.24% | 5.95% |
L2065 | $16.38 | 0.02% | 8.37% |
L2060 | $16.39 | 0.02% | 8.38% |
L2055 | $16.39 | 0.02% | 8.38% |
L2050 | $32.73 | 0.01% | 6.95% |
L2045 | $14.91 | 0.02% | 6.58% |
L2040 | $54.38 | 0.02% | 6.22% |
L2035 | $14.34 | 0.02% | 5.79% |
L2030 | $47.67 | 0.02% | 5.38% |
L2025 | $13.15 | 0.03% | 3.43% |
Linc | $25.61 | 0.03% | 2.82% |