10 Month Modified SMA Method (M-SMA)

Sort by newest Sort by date
jpcavin said:
Shouldn't you remain in "S" until you recover your loses in "S"?:blink:
Click to expand...

In short, yes.

My point was that IF stocks continue to rise, then I should be at about the same point that the method told me to bail to F (even though I went all S the day before - see)

Just curious... :sick:

But it's pretty much too late for today, so I will see where I am at on Monday...
 
RealMoneyIssues said:
In short, yes.

My point was that IF stocks continue to rise, then I should be at about the same point that the method told me to bail to F (even though I went all S the day before - see)

Just curious... :sick:

But it's pretty much too late for today, so I will see where I am at on Monday...
Click to expand...

I played the roulette table today and moved 50% to "I" :sick:
 
RealMoneyIssues said:
Good Luck

I just want to get back on track so I can get back to simple...
Click to expand...

I guess it boils down to if you want to follow the system and just check the signal twice a month as David designed it, or more often, like I think many of us do. I do it daily.

I just plugged in the numbers from the close on the 14, 15, 16 and 17th. Basically this week, here is what the system came up with.

Monday the 14th, Buy S

Tuesday the 15th, Buy F

Wednesday the 16th, Buy F

Thursday the 17th, Buy F

I'd be interested in what todays close does to the system.
 
k0nkuzh0n said:
Yes, but its not an issue of what day you pick for the mid month... Its how the return is calculated. The return is calculated as if you sold on the same day share prices that you got the signal, Which isn't possible with TSP, you have to trade the following day.

So for it to be accurate, you would have to enter the correct date's prices, whether it be the mid month or the end of the month, then if that caused a signal, you'd have to enter the NEXT day's prices and hope the signal is the same. Otherwise the return percentage is off by a day.
Click to expand...


Yes, but isn't that the problem with any system we try and use with our TSP Accounts.

I think it is as good as it can get with our TSP restrictions. It's just another tool we can use to try and earn more for our retirement.

We really won't know how good it is (or isn't) until the end of the year when we see real returns, not back tested numbers.
 
Kaufmanrider said:
Yes, but isn't that the problem with any system we try and use with our TSP Accounts.

I think it is as good as it can get with our TSP restrictions. It's just another tool we can use to try and earn more for our retirement.

We really won't know how good it is (or isn't) until the end of the year when we see real returns, not back tested numbers.
Click to expand...
Yes it is, but that doesn't mean we can't calculate the actual return percentage correctly. Otherwise all the back tested results are semi-bogus because its off by 1 day every time it changes signals.
 
Kaufmanrider said:
OK, when you finish I'd like to see the results.
Click to expand...
haha I'm not going to fix the spreadsheet, I didn't make it in the first place so I don't know how he set it up, and don't really feel like taking the time to examine it. I'm not following this system directly anyways, just using it in conjunction with others.
 
k0nkuzh0n said:
haha I'm not going to fix the spreadsheet, I didn't make it in the first place so I don't know how he set it up, and don't really feel like taking the time to examine it. I'm not following this system directly anyways, just using it in conjunction with others.
Click to expand...

So am I. As slightly off is it might be, I don't think you can accurately back test it with the way the TSP limits our moves. I continue to watch it in real time and as the year continues we can see how it really works.

I did follow it this month, I thought the S fund would continue to climb, it killed me. I should of stuck with I_T's system. I'd still be up 4% for the year instead of under 1%. No IFT's left and I only moved half into F when the System gave the move signal mid month. Still hoping for something from the S.

At least my brokerage (Custodial) accounts are up 2% so far this month following I_T.
 
k0nkuzh0n said:
Yes it is, but that doesn't mean we can't calculate the actual return percentage correctly. Otherwise all the back tested results are semi-bogus because its off by 1 day every time it changes signals.
Click to expand...

You make a great point about it being off by 1 day....I never notice that. I'll fix it and post the new version.

If you take a look at the backtested data, it rarely throws you into the F Fund (3 or 4 times since 2002) and causes whiplash. The other times it puts you in the F Fund in the backtested data, you would be VERY glad it did.

This year's end-of year return for this method probably won't show you anything different than what is already in the backtested data. You will probably earn anywhere from 4-39% this year just as the backtested data shows. Over a period of 10 years, you'll probably average 10-20% per year.
 
Kaufmanrider said:
I should of stuck with I_T's system. I'd still be up 4% for the year instead of under 1%. No IFT's left and I only moved half into F when the System gave the move signal mid month. Still hoping for something from the S.
Click to expand...

It's the nature of any moving average system, you have to account for the lag and the maximum draw-down associated with long-term moving average crossover systems.

You also have to consider did you take the entry at the right time? Had you taken an entry from last September's crossover you'd be up 12.5% from that entry.
 
JTH said:
It's the nature of any moving average system, you have to account for the lag and the maximum draw-down associated with long-term moving average crossover systems.

You also have to consider did you take the entry at the right time? Had you taken an entry from last September's crossover you'd be up 12.5% from that entry.
Click to expand...

Your point about entry is the main reason I decided to stay away from a moving average based system for the time being. If the entry would have occurred back in September, it would be easier to ride out the ups and downs of this type of system. But buying this type of system in January or February may lead to short-term disappointment if the markets make a big move back down to the 10 month moving average. David built in a stop loss which may help to curb the pain of the downside.
 
RealMoneyIssues said:
So, what date are we assessing the prices and what date are we doing the IFT?

Thanks !!!
Click to expand...

This month, on 15 March COB, the assesment and IFT was done. Due to the TSP lag, we were still in S Fund on the 16th then moved to the F Fund after COB on the 16th.

31 March, we will assess and do an IFT. Monday 4 April, we will still be in F Fund due to tsp lag, and will be moved into the S Fund (Assumed) AFTER COB Monday 4 April.
 
RealMoneyIssues said:
I was kinda curious the other day. Where did the 2.92% stop-loss come from? Not the stop-loss per se, but the 2.92% :confused:
Click to expand...

Before adding the stoploss, I noticed in the data it was possible to take heavy losses just before the 1st and 15th, but STILL not be below the SMA at the time of the assessment, thus for another 2 weeks, we would take even more losses. I tried various stoploss percentages (1, 2, 3, 4, 5) and 2.92% was the best stoploss according to the data.
 
Well, it's that time again. The end of the month checkpoint for the M-SMA is giving a BUY signal for the S-Fund and a SELL Signal for the F-Fund. We are going 100% S-Fund. Obviously, we will still be in the F Fund tomorrow due to the TSP lag and our first true day in the S-Find will be Monday the 4th.
 
FWIW, I tried to independently calculate and verify the results of this modified SMA method, and came up with different (lower) numbers. Between 2003 and 2010 I get compound annual returns of 17.4%, with a maximum loss (draw down) of -10.8%.

I also extended the back test to 1988, and from 1988-2010 I calculate compound annual returns of 14.6% with a maximum draw down of -21.6%.

Has anyone else tried to verify this system?
 
You must log in or register to reply here.

Similar threads

TommyIV
September Contest in Progress
Replies
3
Views
4K
TommyIV
TommyIV
tsptalk
  • Article Article
Does 10% ring the bell?
Replies
1
Views
273
JTH
JTH
RevShark
  • Article Article
RevShark's TSP Timing Changes
Replies
1
Views
400
tsptalk
tsptalk
tsptalk
  • Article Article
TSP Talk - The CPI showed inflation under control
Replies
0
Views
268
tsptalk
tsptalk
tsptalk
  • Article Article
TSP Talk - The PPI was good, can the CPI keep cool?
Replies
0
Views
285
tsptalk
tsptalk
Back
Top