WEBVTT

1
00:00:00.060 --> 00:00:00.240
Hello,

2
00:00:00.420 --> 00:00:02.802
welcome back to Papers with Backtest podcast.

3
00:00:03.042 --> 00:00:03.423
Today,

4
00:00:03.623 --> 00:00:06.525
we dive into another algo trading research paper.

5
00:00:06.685 --> 00:00:07.066
We are.

6
00:00:07.206 --> 00:00:07.626
This time,

7
00:00:07.627 --> 00:00:09.648
we're taking a close look at a paper called

8
00:00:10.328 --> 00:00:13.691
Are Return Seasonalities Due to Risk or Mispricing?

9
00:00:14.491 --> 00:00:16.012
Evidence from Seasonal Reversals.

10
00:00:16.373 --> 00:00:16.493
OK,

11
00:00:16.733 --> 00:00:17.394
seasonality.

12
00:00:17.554 --> 00:00:25.440
So we're talking about that thing where certain stocks just seem to perform well or poorly in the same month,

13
00:00:25.620 --> 00:00:26.381
year after year.

14
00:00:26.761 --> 00:00:27.201
Exactly.

15
00:00:27.281 --> 00:00:27.862
And this paper.

16
00:00:28.460 --> 00:00:28.600
Well,

17
00:00:28.601 --> 00:00:30.662
it really digs into the why behind that.

18
00:00:30.722 --> 00:00:30.882
Yeah.

19
00:00:31.002 --> 00:00:33.904
Is it some underlying risk that changes seasonally?

20
00:00:33.944 --> 00:00:35.025
Or is it something else?

21
00:00:35.345 --> 00:00:38.928
Like maybe the market just gets the price wrong temporarily,

22
00:00:39.269 --> 00:00:40.449
some kind of mispricing.

23
00:00:40.590 --> 00:00:41.650
That's the core question.

24
00:00:42.231 --> 00:00:45.173
Are these predictable monthly wiggles driven by,

25
00:00:45.533 --> 00:00:45.774
you know,

26
00:00:45.894 --> 00:00:47.335
fundamental economic risk?

27
00:00:47.815 --> 00:00:50.217
Or are they maybe signals that the market's having a bit of a hiccup?

28
00:00:50.517 --> 00:00:50.818
Right.

29
00:00:50.878 --> 00:00:52.499
And this is where it gets really interesting for us,

30
00:00:52.559 --> 00:00:55.181
especially if you're thinking about actual trading strategies.

31
00:00:55.712 --> 00:00:57.473
Because I guess if it is mispricing,

32
00:00:57.493 --> 00:00:59.555
you'd expect some kind of balancing effect,

33
00:00:59.635 --> 00:01:00.015
wouldn't you?

34
00:01:00.096 --> 00:01:00.216
Well,

35
00:01:00.217 --> 00:01:01.176
that's what the paper argues.

36
00:01:01.236 --> 00:01:02.317
If a stock's price gets,

37
00:01:02.678 --> 00:01:02.818
say,

38
00:01:03.278 --> 00:01:05.320
pushed up too high in January just because it's January,

39
00:01:05.560 --> 00:01:07.581
then maybe its returns in

40
00:01:07.942 --> 00:01:09.002
February, March,

41
00:01:09.323 --> 00:01:12.545
the rest of the year should be a bit lower to compensate.

42
00:01:12.705 --> 00:01:14.347
It can't just stay over-valued forever.

43
00:01:14.627 --> 00:01:14.907
Ah,

44
00:01:15.427 --> 00:01:15.548
OK.

45
00:01:15.668 --> 00:01:17.749
So this idea of seasonal reversals.

46
00:01:18.070 --> 00:01:18.970
That's the term they use.

47
00:01:19.371 --> 00:01:23.674
The thinking is if these monthly patterns are just temporary mispricings.

48
00:01:24.364 --> 00:01:30.569
Then a month where you'd expect high returns should probably be balanced out by months where you'd expect lower returns.

49
00:01:31.470 --> 00:01:32.290
And the other way around,

50
00:01:32.370 --> 00:01:32.611
too,

51
00:01:33.111 --> 00:01:33.751
over the whole year,

52
00:01:33.752 --> 00:01:34.672
it kind of evens out.

53
00:01:34.752 --> 00:01:37.614
And that could be a really valuable insight for traders,

54
00:01:37.654 --> 00:01:38.035
couldn't it?

55
00:01:38.055 --> 00:01:39.396
If it's not just random noise,

56
00:01:39.397 --> 00:01:41.858
but something predictable that corrects itself.

57
00:01:42.238 --> 00:01:42.738
Definitely.

58
00:01:43.099 --> 00:01:44.480
So for this deep dive,

59
00:01:44.720 --> 00:01:49.764
we're really going to zoom in on the trading rules the paper looked at and maybe most importantly,

60
00:01:49.824 --> 00:01:51.185
what the back test showed.

61
00:01:51.385 --> 00:01:51.946
Sounds good.

62
00:01:52.236 --> 00:01:54.298
Let's start with that season reversals concept then.

63
00:01:54.358 --> 00:01:54.598
Okay.

64
00:01:54.718 --> 00:01:57.740
So what they found basically is that stocks that tend to do well in,

65
00:01:57.820 --> 00:01:57.981
say,

66
00:01:58.241 --> 00:01:58.521
April.

67
00:01:58.981 --> 00:01:59.161
Yeah.

68
00:01:59.302 --> 00:02:01.904
Often tend to do less well in the other 11 months.

69
00:02:02.844 --> 00:02:04.025
And the opposite is true too.

70
00:02:04.606 --> 00:02:07.428
Stocks weak in one month might be stronger in others,

71
00:02:07.908 --> 00:02:08.769
like an ebb and flow.

72
00:02:09.129 --> 00:02:09.329
Okay,

73
00:02:09.349 --> 00:02:10.490
a predictable ebb and flow.

74
00:02:11.271 --> 00:02:13.072
So how did they actually measure this?

75
00:02:13.112 --> 00:02:17.676
How did they identify these monthly habits and these reversal effects?

76
00:02:17.956 --> 00:02:18.096
Well,

77
00:02:18.276 --> 00:02:19.517
they used a lot of historical data,

78
00:02:19.617 --> 00:02:19.998
obviously.

79
00:02:20.238 --> 00:02:20.358
Right.

80
00:02:20.378 --> 00:02:21.199
For each stock.

81
00:02:21.532 --> 00:02:24.274
They calculated its average return for January over many,

82
00:02:24.314 --> 00:02:24.775
many years.

83
00:02:24.776 --> 00:02:25.535
Right.

84
00:02:25.536 --> 00:02:29.258
And then they also calculated its average return for all the other months combined,

85
00:02:29.318 --> 00:02:30.399
February through December.

86
00:02:30.979 --> 00:02:31.620
And crucially,

87
00:02:32.020 --> 00:02:35.823
they made sure to skip the most recent year's data when doing these calculations.

88
00:02:36.143 --> 00:02:36.564
Why is that?

89
00:02:37.024 --> 00:02:38.345
To avoid look-ahead bias.

90
00:02:39.086 --> 00:02:46.111
You don't want your strategy simulation to accidentally use information that wouldn't have actually been available at the time you were making the trade decision.

91
00:02:46.211 --> 00:02:46.512
Gotcha.

92
00:02:46.672 --> 00:02:47.132
Makes sense.

93
00:02:47.492 --> 00:02:48.473
Only use past data.

94
00:02:48.773 --> 00:02:48.953
Okay,

95
00:02:49.014 --> 00:02:50.515
so they've identified these tendencies.

96
00:02:50.575 --> 00:02:50.695
Now,

97
00:02:50.852 --> 00:02:52.073
the million dollar question,

98
00:02:53.094 --> 00:02:54.655
did trading on them actually,

99
00:02:54.915 --> 00:02:55.155
you know,

100
00:02:55.235 --> 00:02:55.516
work?

101
00:02:55.696 --> 00:02:56.757
Let's look at the back tests.

102
00:02:57.177 --> 00:02:57.437
First,

103
00:02:57.438 --> 00:03:00.560
they tested a strategy based purely on the same month average return,

104
00:03:00.780 --> 00:03:01.901
the simple seasonality.

105
00:03:01.961 --> 00:03:02.861
They call it ANN.

106
00:03:03.302 --> 00:03:03.782
How did that do?

107
00:03:04.222 --> 00:03:05.023
Surprisingly well,

108
00:03:05.103 --> 00:03:05.403
actually.

109
00:03:05.604 --> 00:03:07.565
The average return was 0.61%

110
00:03:08.005 --> 00:03:08.466
per month.

111
00:03:09.126 --> 00:03:09.366
Okay,

112
00:03:09.367 --> 00:03:10.147
0.61.

113
00:03:10.487 --> 00:03:10.667
Which,

114
00:03:10.868 --> 00:03:11.108
you know,

115
00:03:11.208 --> 00:03:12.909
might not sound earth shattering on its own,

116
00:03:12.949 --> 00:03:15.331
but the T value was 8.37.

117
00:03:15.531 --> 00:03:15.711
Wow,

118
00:03:15.712 --> 00:03:18.434
8.37.

119
00:03:18.435 --> 00:03:20.235
That's statistically very significant,

120
00:03:20.255 --> 00:03:20.555
isn't it?

121
00:03:20.624 --> 00:03:21.445
Extremely.

122
00:03:21.465 --> 00:03:24.707
It strongly suggests this isn't just luck or random chance.

123
00:03:24.967 --> 00:03:27.189
There seems to be a real persistent pattern there.

124
00:03:27.690 --> 00:03:27.810
OK,

125
00:03:27.910 --> 00:03:31.092
so trading on the month's typical performance looks promising.

126
00:03:31.813 --> 00:03:33.614
What about the other side of that coin,

127
00:03:33.654 --> 00:03:34.755
the reversal idea?

128
00:03:35.596 --> 00:03:38.438
Trading based on how a stock does in the other months.

129
00:03:38.858 --> 00:03:40.079
That's the NANN factor.

130
00:03:40.139 --> 00:03:40.620
Exactly.

131
00:03:40.740 --> 00:03:42.341
NANN stands for non-annual.

132
00:03:42.741 --> 00:03:43.202
So here,

133
00:03:43.782 --> 00:03:45.544
the strategy is betting on that reversal.

134
00:03:46.064 --> 00:03:49.266
You buy stocks that have historically done poorly in the other 11 months.

135
00:03:49.456 --> 00:03:51.217
Hoping they'll revert in their good month.

136
00:03:51.358 --> 00:03:51.738
Kind of,

137
00:03:51.798 --> 00:03:54.780
or maybe just identifying stocks whose bad months are particularly bad,

138
00:03:54.840 --> 00:03:56.882
suggesting the good month effect is more pronounced,

139
00:03:57.162 --> 00:03:59.144
and you sell short the ones that do well in the other months.

140
00:03:59.344 --> 00:03:59.524
Right,

141
00:03:59.544 --> 00:04:00.445
the reversal play.

142
00:04:00.665 --> 00:04:01.626
And how did that perform?

143
00:04:02.086 --> 00:04:05.308
That one came in with an average monthly return of 0.45%.

144
00:04:05.769 --> 00:04:06.569
Still positive.

145
00:04:06.669 --> 00:04:07.210
Still positive,

146
00:04:07.250 --> 00:04:07.430
yes.

147
00:04:07.630 --> 00:04:09.632
And the T-value was 4.89.

148
00:04:09.652 --> 00:04:09.852
Okay,

149
00:04:10.012 --> 00:04:11.914
also statistically significant,

150
00:04:11.994 --> 00:04:13.895
though not quite as high as the first one.

151
00:04:14.235 --> 00:04:15.036
Still very solid,

152
00:04:15.076 --> 00:04:15.256
yeah.

153
00:04:16.384 --> 00:04:21.944
It supports the idea that these seasonal highs in one month seem connected to relative lows in the other months.

154
00:04:22.564 --> 00:04:24.584
So both strategies kind of work on their own.

155
00:04:24.624 --> 00:04:25.884
Did they try putting them together?

156
00:04:26.124 --> 00:04:26.324
Sure.

157
00:04:26.325 --> 00:04:30.044
Like looking at the difference between the same month return and the other month return.

158
00:04:30.424 --> 00:04:31.664
That was the AMN factor,

159
00:04:32.064 --> 00:04:33.484
annual minus non-annual?

160
00:04:33.504 --> 00:04:34.124
They did.

161
00:04:34.364 --> 00:04:35.964
And this combined approach,

162
00:04:35.965 --> 00:04:37.524
it gave the best results of the lot.

163
00:04:37.684 --> 00:04:37.924
Really?

164
00:04:38.044 --> 00:04:38.564
How much better?

165
00:04:38.784 --> 00:04:42.744
The average monthly return jumped up to 0.67%.

166
00:04:42.984 --> 00:04:43.224
Okay.

167
00:04:43.364 --> 00:04:44.764
Higher than either individually.

168
00:04:44.804 --> 00:04:44.944
Yep.

169
00:04:45.500 --> 00:04:48.545
And the t-value shot up to 9.93.

170
00:04:48.546 --> 00:04:49.066
Whoa,

171
00:04:49.067 --> 00:04:49.687
nearly 10.

172
00:04:49.727 --> 00:04:50.448
That's huge.

173
00:04:50.768 --> 00:04:51.309
It really is.

174
00:04:51.670 --> 00:04:53.573
Extremely statistically significant.

175
00:04:54.054 --> 00:04:59.182
It strongly suggests that considering both the seasonal tendency and the reversal pattern together,

176
00:04:59.624 --> 00:05:01.345
gives you a much more powerful signal.

177
00:05:01.646 --> 00:05:02.827
That feels intuitive actually.

178
00:05:03.147 --> 00:05:07.070
If you know a stock tends to go up in May and it tends to underperform the rest of the year,

179
00:05:07.490 --> 00:05:09.612
that combination paints a clearer picture maybe.

180
00:05:09.832 --> 00:05:10.412
Seems that way.

181
00:05:10.853 --> 00:05:11.473
And importantly,

182
00:05:11.513 --> 00:05:14.316
the paper notes that neither the basic seasonality factor,

183
00:05:14.496 --> 00:05:14.956
ANN,

184
00:05:15.356 --> 00:05:16.477
nor the reversal factor,

185
00:05:16.637 --> 00:05:17.158
NNN,

186
00:05:17.538 --> 00:05:18.659
fully explains the other.

187
00:05:18.979 --> 00:05:21.221
Meaning they contain some independent information.

188
00:05:21.561 --> 00:05:23.142
They're not just perfectly mirrored images.

189
00:05:23.603 --> 00:05:27.366
Each one captures a slightly different nuance of the expected return pattern.

190
00:05:27.952 --> 00:05:28.332
Interesting.

191
00:05:28.372 --> 00:05:28.553
Okay,

192
00:05:28.573 --> 00:05:32.155
so we have these potentially profitable seasonal strategies.

193
00:05:32.556 --> 00:05:34.577
How do they fit in with the rest of the factors,

194
00:05:34.617 --> 00:05:35.098
you know,

195
00:05:35.418 --> 00:05:36.078
market risk,

196
00:05:36.359 --> 00:05:36.939
size,

197
00:05:37.299 --> 00:05:37.820
value,

198
00:05:37.960 --> 00:05:38.560
momentum,

199
00:05:39.881 --> 00:05:40.882
the usual suspects?

200
00:05:41.002 --> 00:05:42.884
That's a great question for portfolio building.

201
00:05:43.264 --> 00:05:44.285
They check the correlations.

202
00:05:44.605 --> 00:05:48.488
And the combined AMN factor showed pretty low correlations with those traditional factors.

203
00:05:48.668 --> 00:05:49.469
Low correlation.

204
00:05:49.529 --> 00:05:51.170
That's good news for diversification,

205
00:05:51.230 --> 00:05:51.350
right?

206
00:05:51.351 --> 00:05:51.751
Potentially,

207
00:05:51.811 --> 00:05:52.051
yes.

208
00:05:52.651 --> 00:05:54.353
If this seasonal strategy zigs,

209
00:05:54.733 --> 00:05:56.955
when your value or momentum strategy zags,

210
00:05:57.352 --> 00:05:59.212
It could help smooth out your overall ride.

211
00:05:59.392 --> 00:05:59.632
Right.

212
00:05:59.772 --> 00:06:02.232
Uncorrelated returns are often highly sought after.

213
00:06:02.512 --> 00:06:03.912
But did they generate alpha?

214
00:06:04.292 --> 00:06:08.912
Did these strategies produce returns after accounting for exposure to those common risk factors?

215
00:06:08.952 --> 00:06:09.772
They did look at that.

216
00:06:10.112 --> 00:06:13.392
They ran regressions against the Carhartt four-factor model that's market,

217
00:06:13.692 --> 00:06:14.132
size,

218
00:06:14.392 --> 00:06:14.732
value,

219
00:06:14.733 --> 00:06:15.232
and momentum.

220
00:06:15.272 --> 00:06:15.892
And the result?

221
00:06:15.992 --> 00:06:17.352
Significant alpha across the board.

222
00:06:18.212 --> 00:06:19.572
The simple seasonality factor,

223
00:06:19.772 --> 00:06:20.172
ANN,

224
00:06:20.532 --> 00:06:22.052
had an alpha of 0.64%

225
00:06:22.312 --> 00:06:22.732
per month.

226
00:06:23.112 --> 00:06:23.492
T-STAT,

227
00:06:23.493 --> 00:06:25.572
8.79.

228
00:06:25.573 --> 00:06:26.532
The reversal factor.

229
00:06:26.884 --> 00:06:29.906
and A&M had an alpha of 0.35%

230
00:06:30.027 --> 00:06:30.407
per month,

231
00:06:31.107 --> 00:06:32.829
T-stat 6.17.

232
00:06:32.830 --> 00:06:33.669
Still significant.

233
00:06:34.030 --> 00:06:34.570
Very much so.

234
00:06:35.211 --> 00:06:38.593
And the combined A&M factor delivered an alpha of 0.66%

235
00:06:38.613 --> 00:06:38.994
per month,

236
00:06:39.454 --> 00:06:41.896
with a T-stat of 9.70.

237
00:06:41.936 --> 00:06:45.379
Almost identical alpha to the combined return and that massive T-stat again.

238
00:06:45.579 --> 00:06:46.059
Exactly.

239
00:06:46.559 --> 00:06:49.522
It means that even after you account for the returns you'd expect,

240
00:06:49.902 --> 00:06:51.944
just from being exposed to market movements,

241
00:06:52.044 --> 00:06:52.704
small caps,

242
00:06:52.784 --> 00:06:53.505
value stocks,

243
00:06:53.525 --> 00:06:54.526
or momentum stocks,

244
00:06:54.972 --> 00:06:58.512
These seasonal strategies still generated significant excess returns.

245
00:06:58.792 --> 00:07:00.872
That really strengthens the case for mispricing,

246
00:07:00.912 --> 00:07:01.292
doesn't it?

247
00:07:01.332 --> 00:07:02.492
If it was just risk,

248
00:07:02.632 --> 00:07:05.152
the standard factors should have explained more of it away.

249
00:07:05.312 --> 00:07:07.412
That's certainly how the evidence seems to lean.

250
00:07:07.852 --> 00:07:10.852
It suggests these patterns aren't just capturing known risk premiums.

251
00:07:11.192 --> 00:07:11.312
Now,

252
00:07:11.313 --> 00:07:12.252
you mentioned reversals.

253
00:07:12.512 --> 00:07:14.512
How does the seasonal reversal compare to,

254
00:07:14.652 --> 00:07:14.892
like,

255
00:07:15.132 --> 00:07:17.172
standard long-term reversal strategies,

256
00:07:17.173 --> 00:07:17.372
you know,

257
00:07:17.392 --> 00:07:20.232
where docs that have been beaten down for years tend to bounce back?

258
00:07:20.432 --> 00:07:20.872
Good point.

259
00:07:20.972 --> 00:07:21.792
They did compare it.

260
00:07:22.132 --> 00:07:24.192
They looked at a typical long-term reversal factor.

261
00:07:24.440 --> 00:07:24.860
LTEV.

262
00:07:25.461 --> 00:07:29.364
The long term reversal factor had a lower average return about 0.29 percent per month.

263
00:07:29.524 --> 00:07:31.345
T-stat around 2.95.

264
00:07:31.526 --> 00:07:31.646
OK.

265
00:07:31.866 --> 00:07:32.767
Lower returns,

266
00:07:32.927 --> 00:07:35.529
less statistical significance compared to the seasonal one.

267
00:07:35.669 --> 00:07:35.889
Right.

268
00:07:35.989 --> 00:07:36.810
And more importantly,

269
00:07:36.850 --> 00:07:43.295
when they regressed LTEV on just the Fama French three factor model market size value,

270
00:07:44.075 --> 00:07:46.277
its alpha wasn't statistically significant.

271
00:07:46.818 --> 00:07:47.218
Ah.

272
00:07:47.678 --> 00:07:51.862
So long term reversal seems largely explained by standard factor.

273
00:07:52.112 --> 00:07:54.034
Whereas these seasonal reversals are not.

274
00:07:55.215 --> 00:07:56.155
They're distinct phenomena.

275
00:07:56.696 --> 00:08:01.039
The seasonal ups and downs aren't just a mini version of long term mean reversion.

276
00:08:01.439 --> 00:08:04.442
So it looks like these seasonal factors offer something genuinely different.

277
00:08:04.822 --> 00:08:09.746
Did the paper touch on what this might mean for overall portfolio performance like risk adjusted returns?

278
00:08:09.946 --> 00:08:10.266
It did.

279
00:08:10.306 --> 00:08:13.048
They did some analysis looking at maximum sharp ratios.

280
00:08:13.309 --> 00:08:15.010
The measure of risk adjusted returns.

281
00:08:15.070 --> 00:08:15.490
Exactly.

282
00:08:16.424 --> 00:08:19.006
The results suggested that adding these seasonal factors,

283
00:08:19.066 --> 00:08:22.789
particularly the combined AMN factor to a portfolio of traditional factors,

284
00:08:23.209 --> 00:08:26.952
could potentially lead to a noticeable improvement in the overall Sharpe ratio.

285
00:08:27.333 --> 00:08:29.114
So better bang for your buck risk wise.

286
00:08:29.174 --> 00:08:29.975
That's the implication.

287
00:08:30.035 --> 00:08:30.175
Yeah.

288
00:08:30.535 --> 00:08:31.996
Improved risk adjusted performance.

289
00:08:32.336 --> 00:08:32.457
OK,

290
00:08:32.797 --> 00:08:34.598
let's try to boil this down then.

291
00:08:35.259 --> 00:08:39.862
The research presents pretty compelling evidence for these predictable seasonal reversals.

292
00:08:40.703 --> 00:08:42.204
Stocks strong in one month,

293
00:08:42.484 --> 00:08:45.587
often offset by weakness in others and vice versa.

294
00:08:45.972 --> 00:08:50.672
And it seems more likely linked to temporary mispricing than just shifting risk.

295
00:08:50.772 --> 00:08:51.792
That's the main thrust,

296
00:08:51.872 --> 00:08:52.052
yes.

297
00:08:52.252 --> 00:08:53.952
And the trading strategies built on this,

298
00:08:54.052 --> 00:08:56.452
especially that combined AMN factor,

299
00:08:56.952 --> 00:08:59.572
showed really strong backtest results,

300
00:09:00.132 --> 00:09:06.432
significant returns and crucially significant alpha even after accounting for standard factors.

301
00:09:06.433 --> 00:09:06.972
Precisely.

302
00:09:07.092 --> 00:09:07.812
The numbers,

303
00:09:08.052 --> 00:09:11.452
particularly the T-stats and the alpha results are quite persuasive.

304
00:09:11.672 --> 00:09:12.652
So for you listening in,

305
00:09:12.653 --> 00:09:14.092
this definitely gives some food for thought.

306
00:09:14.556 --> 00:09:16.896
How might these recurring seasonal glitches,

307
00:09:16.956 --> 00:09:17.896
these mispricings,

308
00:09:18.276 --> 00:09:21.516
fit with other market anomalies you track or even your own strategies?

309
00:09:21.576 --> 00:09:21.816
Right.

310
00:09:21.896 --> 00:09:25.196
Could weaving in a seasonal perspective actually give you an additional edge?

311
00:09:25.656 --> 00:09:27.096
It's definitely something worth mulling over.

312
00:09:27.236 --> 00:09:30.796
It seems like paying attention to the calendar might be more important than some people think.

313
00:09:31.236 --> 00:09:34.776
This research certainly suggests it could be a fruitful area to explore,

314
00:09:35.056 --> 00:09:38.376
looking beyond just the usual factors for potential opportunities.

315
00:09:38.776 --> 00:09:41.376
Thank you for tuning in to Papers with Backtest podcast.

316
00:09:41.576 --> 00:09:43.996
We hope today's episode gave you useful insights.

317
00:09:44.384 --> 00:09:46.491
Join us next time as we break down more research.

318
00:09:46.631 --> 00:09:47.915
And for more papers and backtests,

319
00:09:47.935 --> 00:09:50.945
find us at HTPS.PapersWithBacktests.com.

320
00:09:51.506 --> 00:09:52.008
Happy trading.