Re: Unemployment Rate over 12 Month MA Strategy
OK, here are the new 5, 10 & 15 year annualized returns:
5 Year Annualized Returns
5 Year |
G Fund |
F Fund |
C Fund |
S Fund |
I Fund |
UR C&F |
UR S&F |
CAGR |
2.03% |
3.57% |
12.63% |
10.65% |
3.78% |
12.63% |
10.65% |
Risk |
0.34% |
3.57% |
11.31% |
15.58% |
13.39% |
11.31% |
15.58% |
10 Year Annualized ReturnsCAGR |
2.94% |
4.74% |
7.36% |
8.03% |
3.20% |
12.54% |
12.25% |
Risk |
1.14% |
2.86% |
18.03% |
21.42% |
20.67% |
9.21% |
12.69% |
15 Year Annualized Returns
CAGR |
3.51% |
5.12% |
5.03% |
7.61% |
3.56% |
10.28% |
10.81% |
Risk |
1.26% |
2.95% |
18.22% |
21.44% |
21.31% |
8.55% |
11.29% |
These results look about the same to a little better than the previous ones. It looks like it helped reducing the whipsawing.
Re: Unemployment Rate over 12 Month MA Strategy
Now here is something interesting that I don't think would impact the Unemployment numbers: Verizon Strike May Depress U.S. Jobs Report, Labor Data Show - Bloomberg
Interesting to get information like this real time. If this affected our numbers at all I would think it would only amount to one of those false spikes we occasionally see in our data.
Re: Unemployment Rate over 12 Month MA Strategy
The Unemployment rate for May was released today. It is now down to 4.7%. That's a significant drop from the 5.0% of the last couple of months. :suspicious: Let's see if & when this figure gets revised, shall we? This is an election year after all. :rolleyes:
Rate: |
4.7% |
12 MMA: |
5.03% |
Trend: |
Still going down |
Re: Unemployment Rate over 12 Month MA Strategy
Here is the unemployment rate for June:
Rate: |
4.9% |
12 MMA: |
4.99% |
Trend: |
Still going down. |
Re: Unemployment Rate over 12 Month MA Strategy
Here is the unemployment rate for July:
Rate: |
4.9% |
12 MMA: |
4.96% |
Crossover: |
rate still below 12 MMA |
Trend: |
still going down. |
That means we are still in the market.
Re: Unemployment Rate over 12 Month MA Strategy
Here is the unemployment rate for August:
Rate: |
4.9% |
12 MMA: |
4.94% |
Crossover: |
rate still below 12 MMA |
Trend: |
still going down |
That means we are still in the market.
Re: Unemployment Rate over 12 Month MA Strategy
The Unemployment rate for September was released this morning. It poped up to 5%. That's a slight increase from the 4.9% of the last couple of months. The trend is still going down, though. That means you have a decision to make. :suspicious:
Rate: |
5.0% |
12 MMA: |
4.93% |
Crossover: |
rate crossed above the 12 MMA |
Trend: |
still going down |
The original trigger, as described in the article linked by Tom4Jean, would have us head for safety now (get out of the market) because the Unemployment has crossed over it's 12 month moving average. If you want to use this trigger you would get out before November and stay out until it crosses under it's 12 MMA again. Things to consider are that our analysis showed that 1) this trigger tended to precede economic downturns by several months and 2) it also produced some one month spikes that whipsawed us in/out of the market way before an actual change in economic conditions occurred.
The alternate trigger, as described in the article linked by Tsunami, would have you stay in for now because the trend of the 12 MMA is still going down. Things to consider for this trigger are 1) following the trend tends to smooth out the spikes so you don't get so many whipsaws and 2) it still precedes the actual economic downturn but not by as great a margin as with the first trigger.
So what do you do now? Looking at the spreadsheet it appears that for an actual economic down turn the crossover trigger typically precedes the trend trigger, usually by a month, and they both tend to precece an actuall downturn. My inclination at this point is to wait until the jobs report for next month to see if this is a one month spike or if the trend changes as well. For the trend to change next month the unemployment rate would have to rise to 5.1%. At 5.0% the trend would still be down while the crossover is up. If both triggers have changed by next month I would certainly be looking to head to safety before next year.
Re: Unemployment Rate over 12 Month MA Strategy
Quote:
Originally Posted by
Cactus
If both triggers have changed by next month I would certainly be looking to head to safety before next year.
That is pretty much what the guy said who's blog I read:
Tactical asset allocation
Re: Unemployment Rate over 12 Month MA Strategy
Here is the unemployment rate for October:
RATE: |
4.9% |
12 MMA: |
4.93% |
Crossover: |
Rate crossed back below the 12 MMA |
Trend: |
Leveling out |
Looks like last month's crossover was indeed a one month spike that would have whipsawed you out and now back in the market if you had followed it.
The place to be for now is in the market according to both the Trend and Crossover triggers.
Re: Unemployment Rate over 12 Month MA Strategy
Here is the unemployment rate for November:
Rate: |
4.6% |
12 MMA: |
4.89% |
Crossover: |
rate still below 12 MMA |
Trend: |
going down |
That means we are still in the market.
Re: Unemployment Rate over 12 Month MA Strategy
Here is the unemployment rate for December:
Rate: |
4.7% |
12 MMA: |
4.87% |
Crossover: |
rate still below 12 MMA |
Trend: |
still going down |
That means we are still in the market
Re: Unemployment Rate over 12 Month MA Strategy
I checked https://data.bls.gov/timeseries/LNS14000000 again this morning and noticed that the September and October numbers had been revised from what had originally been report as follows:
Sep: 5.0 -> 4.9
Oct: 4.9 -> 4.8
Using these revised numbers eliminates the whipsawing we experienced at the time. Unfortunately we didn't have the benefit of those revised numbers back then. Doing this real time requires that we use the numbers as first reported. Unfortunately this also calls into question the backtesting analysis that was performed on this method since all the historical data would have been corrected. I was afraid of that at the time and it appears those concerns are well founded.
It seams the only thing we can do now is keep going forward real time with the uncorrected data and see if that gives us similar results to our backtested results. That's what I plan to do here anyways. Let me know if anyone has any better ideas.