This topic contains 5 replies, has 2 voices, and was last updated by tmj 3 years, 4 months ago.
November 11, 2015 at 3:20 pm #17556
Has anyone created a monthly allocations model or formula for TSP?
I have been working on one for a a couple of months and am trying to see if others have done this.
I’ve created a website to memorialize my allocations and eventually prove its worth –> http://www.TSPinvestor.com
To be clear, this system has no human decision making on the monthly allocations .. and that is what I’m wondering about that others have developed.
I ran historical data through the model and got a 2005-2014 return for a Basic version at 12%. This uses only the fund historical data for input. And the Plus version returns 19%, and uses a few leading economic factors as input in addition to the fund historical data.
I’m interested in seeing what others have done. I’ve seen plenty of monthly allocation advice, but I’m not convinced it doesn’t involve a human decision.
www.TSPinvestor.comNovember 13, 2015 at 4:19 pm #17561
When something sounds to good to be true…November 13, 2015 at 5:52 pm #17562
not sure what that means .. and that isn’t an answer to my question .. if you have some useful comments and questions, I’m all ears
www.TSPinvestor.comNovember 21, 2015 at 8:49 pm #17651
I cannot tell exactly what you are doing based on your website but most Prospectus state in some manner: Past Performance is Not Necessarily Indicative of Future Results. Investors are often not logical and can often be irrational. I think both options for November are too high in G & F. If you look at the allocations for Feb 2015, your Plus puts everything in G & F when the C, S & I each gain about 6% for the month. In August, you have 92% in C, S & I in the plus model, each losing between 5.8 and 7.4% if you follow your rules. Not sure how you can make your investment grow this way. From 09 Jul 2004 to 20 Nov 2015, the S&P 500 gained 85.64%; the Dow Jones U.S. Completion Total Stock Market Index gained 133.28% So I believe putting it all in C or S would have worked better over the same time period.
Barclays Aggregate Bond Fund in comparison was 9.18% over the same period, not including impact of monthly dividends.
Assuming you allocated either of your recommendations based on your 31 Oct predictions on a balance of $100K, as of 2 Nov (when the change would be made as of the close of business in TSP), you would currently have $99,953.68 under the Basic and $99,196.34 under the Plus as of Friday, 20 Nov. If contributions were made on 6 & 20 November in the same allocations, the contributions on the 6th would have resulted in a slight gain under the Basic and loss under Plus for those shares. I think you need to do some more analysis going forward, but we still have over a week to go this month but I’m expecting the stock funds to increase more than either G or F. Good LuckDecember 14, 2015 at 10:38 am #17716
I am using the Plus as my strategy. But have only started recently. I am really in the process of proving this model works and am open to all critique and suggestions. As the old saying goes on models – all models are wrong; some are useful. This strategy depends primarily upon gauging tendencies, not exact predictions of returns. Is it accurate? The model performance using historical data suggests some utility. Is it good for me, as TSP participant? The answer thus far is yes. To answer that I have to compare my old strategy to the current. I had a 50% C and 50% S buy/hold/reallocate yearly or so, for more than 12 years. And thus far the Plus has been better – though the data points are limited.
So ….. we’ll see.
www.TSPinvestor.comDecember 22, 2015 at 2:12 pm #17803
Thanks for sharing your insights on a new potential model through your website.
Are you planning on sharing the spreadsheet too or just the model’s recommendations?
Additionally, are the fund transfer days dynamic based on external stimulus like MACD or static such as last day of month?
You must be logged in to reply to this topic.