WEBVTT

1
00:00:00.040 --> 00:00:00.180
Hello,

2
00:00:00.280 --> 00:00:02.100
welcome back to Papers with Backtest podcast.

3
00:00:02.141 --> 00:00:04.781
Today we dive into another Algo trading research paper.

4
00:00:05.160 --> 00:00:05.641
Hi there.

5
00:00:05.801 --> 00:00:05.941
Yeah,

6
00:00:05.980 --> 00:00:13.125
today we're looking at a really interesting one called Alpha Momentum in Country and Industry Equity Indexes.

7
00:00:13.586 --> 00:00:14.781
It's by Zaremba,

8
00:00:15.180 --> 00:00:15.586
Umutlu,

9
00:00:16.047 --> 00:00:17.281
and Karathanisopoulos.

10
00:00:17.625 --> 00:00:17.828
Okay,

11
00:00:17.984 --> 00:00:18.797
Alpha Momentum.

12
00:00:18.844 --> 00:00:19.938
So unpack that for us.

13
00:00:20.484 --> 00:00:23.031
What's the central question they're tackling here?

14
00:00:23.438 --> 00:00:23.578
Well,

15
00:00:23.766 --> 00:00:28.641
they're essentially asking if a country or an industry has performed really well in the past,

16
00:00:28.768 --> 00:00:30.469
And even after you adjust for risk,

17
00:00:30.950 --> 00:00:31.772
that's the alpha part.

18
00:00:31.791 --> 00:00:36.135
Does that predict it'll keep doing well or does it mean it's maybe due for a fall?

19
00:00:36.377 --> 00:00:36.577
Right.

20
00:00:36.674 --> 00:00:37.620
Past performance,

21
00:00:37.838 --> 00:00:41.057
but specifically risk adjusted performance predicting the future.

22
00:00:41.338 --> 00:00:42.979
And they used a pretty big data set for this,

23
00:00:43.018 --> 00:00:43.424
I gather.

24
00:00:43.565 --> 00:00:43.706
Oh,

25
00:00:43.846 --> 00:00:44.682
absolutely massive.

26
00:00:44.706 --> 00:00:46.362
They looked at equity indexes from,

27
00:00:46.440 --> 00:00:46.768
get this,

28
00:00:46.909 --> 00:00:48.252
51 stock markets.

29
00:00:48.690 --> 00:00:53.252
And the data goes from 1973 all the way to 2018.

30
00:00:53.253 --> 00:00:53.315
Wow.

31
00:00:53.316 --> 00:00:53.393
Okay.

32
00:00:53.394 --> 00:00:54.018
And on top of that,

33
00:00:54.019 --> 00:00:56.409
887 industry indexes.

34
00:00:56.502 --> 00:00:56.831
So yeah,

35
00:00:57.221 --> 00:00:58.081
really comprehensive.

36
00:00:58.264 --> 00:00:59.766
That's some serious data crunching.

37
00:01:00.427 --> 00:01:02.329
So what were the headline findings?

38
00:01:02.368 --> 00:01:04.731
What should we be paying attention to from all that analysis?

39
00:01:05.071 --> 00:01:06.431
They found two main patterns,

40
00:01:06.470 --> 00:01:06.649
really.

41
00:01:06.735 --> 00:01:07.071
First,

42
00:01:07.149 --> 00:01:07.852
in the short term,

43
00:01:07.892 --> 00:01:09.938
there's this thing they call alpha momentum.

44
00:01:10.798 --> 00:01:11.259
Basically,

45
00:01:11.556 --> 00:01:14.079
if a market has shown strong risk adjusted returns,

46
00:01:14.118 --> 00:01:14.360
you know,

47
00:01:14.876 --> 00:01:16.188
strong alpha recently,

48
00:01:16.657 --> 00:01:18.907
it tends to keep outperforming for a bit longer,

49
00:01:19.001 --> 00:01:20.063
kind of like a hot streak,

50
00:01:20.501 --> 00:01:22.360
but measured properly for risk.

51
00:01:22.532 --> 00:01:22.704
Okay,

52
00:01:22.735 --> 00:01:24.204
that makes a certain intuitive sense.

53
00:01:24.704 --> 00:01:25.157
Momentum,

54
00:01:25.376 --> 00:01:25.954
but for Alpha,

55
00:01:26.376 --> 00:01:27.813
what was the second pattern?

56
00:01:28.024 --> 00:01:31.308
The second one is kind of the opposite over the longer term.

57
00:01:31.366 --> 00:01:32.827
They call it alpha reversal.

58
00:01:32.929 --> 00:01:33.351
Reversal.

59
00:01:33.390 --> 00:01:33.628
Okay.

60
00:01:33.730 --> 00:01:33.972
Yeah.

61
00:01:34.151 --> 00:01:39.097
So countries or industries that have had really high alpha over a longer period,

62
00:01:39.175 --> 00:01:40.480
say several years,

63
00:01:41.276 --> 00:01:41.394
well,

64
00:01:41.395 --> 00:01:43.081
they actually tend to underperform later on.

65
00:01:43.097 --> 00:01:45.464
It's like they get overextended and then snap back.

66
00:01:45.589 --> 00:01:45.761
Huh.

67
00:01:46.526 --> 00:01:47.761
Short-term continuation,

68
00:01:47.901 --> 00:01:50.511
long-term mean reversion based on alpha.

69
00:01:51.386 --> 00:01:55.089
And the paper suggests you could actually trade based on this,

70
00:01:55.151 --> 00:01:55.933
build strategies.

71
00:01:56.200 --> 00:01:56.661
Exactly.

72
00:01:56.721 --> 00:01:57.922
That's the practical angle.

73
00:01:57.962 --> 00:02:00.404
The idea is you could potentially use these patterns for,

74
00:02:00.486 --> 00:02:00.666
say,

75
00:02:01.103 --> 00:02:02.506
international equity allocation,

76
00:02:02.826 --> 00:02:06.674
knowing which markets might be poised to rise or fall based on their past alpha.

77
00:02:06.994 --> 00:02:07.111
OK,

78
00:02:07.252 --> 00:02:08.689
let's get into the nuts and bolts then.

79
00:02:09.174 --> 00:02:10.838
How did they actually measure this alpha?

80
00:02:11.072 --> 00:02:11.978
It sounds crucial.

81
00:02:12.252 --> 00:02:12.517
Right.

82
00:02:12.642 --> 00:02:13.955
They didn't just pick one method.

83
00:02:14.064 --> 00:02:16.955
They used four different pretty standard factor models,

84
00:02:17.033 --> 00:02:17.283
you know,

85
00:02:17.299 --> 00:02:19.846
ways to strip out market wide effects and see what's left.

86
00:02:20.017 --> 00:02:20.721
Like CAPM.

87
00:02:20.971 --> 00:02:21.080
Yep.

88
00:02:21.158 --> 00:02:21.939
CAPM was one.

89
00:02:22.484 --> 00:02:26.407
Then the Fama-French three-factor model that adds size and value factors.

90
00:02:26.567 --> 00:02:28.009
Also the Carhartt four-factor,

91
00:02:28.110 --> 00:02:29.751
which brings in momentum itself.

92
00:02:29.891 --> 00:02:30.110
Okay.

93
00:02:30.395 --> 00:02:32.512
And another three-factor model from Asness,

94
00:02:32.677 --> 00:02:33.098
Frasini,

95
00:02:33.153 --> 00:02:33.716
and Peterson.

96
00:02:34.294 --> 00:02:35.497
And a key thing,

97
00:02:36.020 --> 00:02:38.661
they scaled the calculated alpha by volatility.

98
00:02:39.145 --> 00:02:39.442
Ah,

99
00:02:39.927 --> 00:02:41.130
volatility scaling.

100
00:02:41.661 --> 00:02:45.020
So looking at the consistency of that excess return,

101
00:02:45.114 --> 00:02:46.192
not just the raw number,

102
00:02:46.708 --> 00:02:48.536
makes sense for comparing diverse markets.

103
00:02:48.630 --> 00:02:49.130
Precisely.

104
00:02:49.348 --> 00:02:49.973
A smoother,

105
00:02:50.192 --> 00:02:51.848
more comparable alpha signal.

106
00:02:52.036 --> 00:02:54.719
So with this volatility adjusted alpha calculated,

107
00:02:55.018 --> 00:02:56.983
how did they build the actual trading rules?

108
00:02:57.139 --> 00:02:58.241
Let's start with the momentum one,

109
00:02:58.522 --> 00:02:58.803
AM.

110
00:02:59.002 --> 00:02:59.225
Okay,

111
00:02:59.643 --> 00:03:00.022
AM.

112
00:03:00.545 --> 00:03:03.389
That uses the alpha calculated over the trailing 12 months,

113
00:03:03.428 --> 00:03:04.428
but they lagged it slightly.

114
00:03:04.545 --> 00:03:06.374
So months T12 back to T1.

115
00:03:06.811 --> 00:03:07.053
Right,

116
00:03:07.131 --> 00:03:08.217
skipping the very last month.

117
00:03:08.530 --> 00:03:08.749
Yeah.

118
00:03:09.108 --> 00:03:11.592
And the strategy was a classic long-short approach.

119
00:03:12.014 --> 00:03:16.264
They ranked all the countries or all the industries based on this recent alpha momentum.

120
00:03:16.545 --> 00:03:16.936
And then?

121
00:03:17.139 --> 00:03:19.217
They went long the top 20%.

122
00:03:19.316 --> 00:03:22.519
the ones with the highest AMOM and shorted the bottom 20%,

123
00:03:22.520 --> 00:03:23.562
the ones with the lowest AMO.

124
00:03:24.159 --> 00:03:26.444
Betting on the winners continuing and the losers lagging.

125
00:03:26.741 --> 00:03:27.163
Makes sense.

126
00:03:27.663 --> 00:03:30.108
So how did that perform in their back tests?

127
00:03:30.483 --> 00:03:33.733
Did this AMOM strategy actually generate alpha itself?

128
00:03:33.991 --> 00:03:34.452
It did.

129
00:03:34.468 --> 00:03:35.937
And the results were pretty striking,

130
00:03:36.015 --> 00:03:37.968
especially for the equal weighted portfolios.

131
00:03:38.327 --> 00:03:40.655
For countries using the three factor model alpha,

132
00:03:40.874 --> 00:03:44.405
they reported an average monthly alpha of 0.92%.

133
00:03:44.499 --> 00:03:45.890
0.92%

134
00:03:45.891 --> 00:03:46.249
per month.

135
00:03:46.265 --> 00:03:46.921
That's significant.

136
00:03:46.968 --> 00:03:47.108
Yeah.

137
00:03:47.249 --> 00:03:48.765
With a sharp ratio of 0.58.

138
00:03:48.856 --> 00:03:49.677
And for industries,

139
00:03:49.678 --> 00:03:50.378
it was even better.

140
00:03:50.698 --> 00:03:52.480
Alpha of 1.41%

141
00:03:52.481 --> 00:03:52.839
per month,

142
00:03:52.921 --> 00:03:54.980
sharp ratio of 0.93.

143
00:03:54.981 --> 00:03:55.101
Wow,

144
00:03:55.300 --> 00:03:55.921
nearly 1%

145
00:03:56.242 --> 00:03:57.441
sharp for the industry strategy.

146
00:03:57.683 --> 00:03:58.722
That's impressive on paper.

147
00:03:59.347 --> 00:04:00.871
But what about real world friction,

148
00:04:01.050 --> 00:04:02.027
like trading costs?

149
00:04:02.152 --> 00:04:02.793
Good question.

150
00:04:02.972 --> 00:04:03.691
They looked into that.

151
00:04:03.949 --> 00:04:06.035
The MM strategy held up surprisingly well.

152
00:04:06.207 --> 00:04:06.847
For countries,

153
00:04:06.910 --> 00:04:07.832
it was profitable,

154
00:04:07.972 --> 00:04:10.332
even assuming one-way trading costs of up to 1%.

155
00:04:10.675 --> 00:04:10.832
1%,

156
00:04:11.175 --> 00:04:11.879
that's quite a buffer.

157
00:04:12.238 --> 00:04:12.566
It is.

158
00:04:12.816 --> 00:04:15.347
And industries could handle even higher costs.

159
00:04:15.738 --> 00:04:16.066
Plus,

160
00:04:16.175 --> 00:04:17.972
the effect wasn't super short-lived.

161
00:04:18.420 --> 00:04:20.902
It were make profitable even if you held the positions for,

162
00:04:21.062 --> 00:04:21.243
say,

163
00:04:21.363 --> 00:04:22.144
10 or 12 months.

164
00:04:22.204 --> 00:04:22.323
OK,

165
00:04:22.706 --> 00:04:23.183
robust.

166
00:04:23.886 --> 00:04:25.706
Did it work better in some markets than others?

167
00:04:26.249 --> 00:04:27.652
Like small versus large?

168
00:04:27.886 --> 00:04:29.792
That's where the evaluated results come in.

169
00:04:30.175 --> 00:04:32.214
When they weighted portfolios by market size,

170
00:04:32.215 --> 00:04:33.316
the alphas were lower,

171
00:04:33.816 --> 00:04:34.472
which suggests,

172
00:04:34.519 --> 00:04:34.722
yeah,

173
00:04:34.753 --> 00:04:38.159
the effect is maybe weaker or harder to capture in those bigger,

174
00:04:38.488 --> 00:04:40.066
perhaps more efficient markets.

175
00:04:40.128 --> 00:04:40.597
Interesting.

176
00:04:40.659 --> 00:04:43.081
So maybe more juice to squeeze in the smaller pots.

177
00:04:43.503 --> 00:04:43.628
OK,

178
00:04:43.722 --> 00:04:44.800
let's flip to the other strategy,

179
00:04:44.847 --> 00:04:45.613
the reversal one.

180
00:04:45.831 --> 00:04:46.269
AREV,

181
00:04:46.425 --> 00:04:46.909
you called it.

182
00:04:46.941 --> 00:04:47.066
Yep.

183
00:04:47.590 --> 00:04:49.753
ARRIV for alpha reversal.

184
00:04:49.952 --> 00:04:52.577
This one looked at alpha over a much longer window,

185
00:04:52.854 --> 00:04:53.456
60 months,

186
00:04:53.557 --> 00:04:54.296
so five years,

187
00:04:54.698 --> 00:04:55.018
but again,

188
00:04:55.096 --> 00:04:56.178
lagging the most recent year.

189
00:04:56.241 --> 00:04:58.124
So months T60 down to T13.

190
00:04:58.741 --> 00:04:59.944
Five years of alpha history,

191
00:05:00.046 --> 00:05:00.944
skipping the last one.

192
00:05:01.202 --> 00:05:01.382
Yeah.

193
00:05:01.421 --> 00:05:03.561
And the strategy here is the opposite of momentum.

194
00:05:03.780 --> 00:05:04.280
Exactly.

195
00:05:04.389 --> 00:05:05.546
Betting on mean reversion.

196
00:05:05.921 --> 00:05:07.342
They went long the bottom 20%

197
00:05:07.405 --> 00:05:11.421
of the countries or industries with the lowest long-term alpha and shorted the top 20%,

198
00:05:11.780 --> 00:05:13.233
the ones with the highest long-term alpha.

199
00:05:13.530 --> 00:05:17.217
So buying the long-term alpha losers and selling the long-term alpha winners.

200
00:05:17.642 --> 00:05:20.505
What did the backtests show for AREV?

201
00:05:20.985 --> 00:05:21.146
Well,

202
00:05:21.206 --> 00:05:22.628
as you'd expect of reversal holds,

203
00:05:22.629 --> 00:05:24.569
the strategy generated negative alpha.

204
00:05:25.030 --> 00:05:28.190
For the equal-weighted country portfolios using the three-factor model,

205
00:05:28.495 --> 00:05:30.995
the alpha was negative 0.52%

206
00:05:31.019 --> 00:05:31.792
per month.

207
00:05:31.979 --> 00:05:33.440
Negative 0.52.

208
00:05:33.456 --> 00:05:33.620
Okay.

209
00:05:33.792 --> 00:05:34.917
And for industries,

210
00:05:35.183 --> 00:05:35.667
similarly,

211
00:05:35.886 --> 00:05:37.542
negative 0.49%

212
00:05:37.543 --> 00:05:37.901
per month.

213
00:05:38.120 --> 00:05:39.417
The Sharpe ratios were lower,

214
00:05:39.511 --> 00:05:39.745
too,

215
00:05:39.901 --> 00:05:41.370
compared to the AOM strategy.

216
00:05:41.573 --> 00:05:42.792
So evidence for reversal,

217
00:05:42.886 --> 00:05:46.558
but perhaps not as strong or consistent as the short-term momentum effect.

218
00:05:46.774 --> 00:05:47.796
That seems to be the case.

219
00:05:47.995 --> 00:05:48.395
And again,

220
00:05:48.737 --> 00:05:51.139
when they looked at value-weighted portfolios for AREV,

221
00:05:51.421 --> 00:05:52.960
the results kind of washed out.

222
00:05:53.022 --> 00:05:56.421
They became statistically insignificant for both countries and industries.

223
00:05:56.561 --> 00:06:02.350
So both alpha momentum and alpha reversal seem stronger or more exploitable in equal-weighted universes,

224
00:06:02.553 --> 00:06:05.710
perhaps hinting at effects in smaller or less efficient markets.

225
00:06:05.928 --> 00:06:07.756
That definitely seems to be a recurring theme,

226
00:06:07.757 --> 00:06:08.022
yes.

227
00:06:08.272 --> 00:06:08.397
Now,

228
00:06:08.553 --> 00:06:12.522
one really interesting comparison was alpha momentum versus plain old price momentum.

229
00:06:12.678 --> 00:06:13.741
How did those stack up?

230
00:06:13.913 --> 00:06:14.960
This was a key finding.

231
00:06:15.106 --> 00:06:18.230
They found that alpha momentum basically subsumes price momentum.

232
00:06:18.589 --> 00:06:19.230
Subsumes,

233
00:06:19.472 --> 00:06:19.730
meaning?

234
00:06:20.113 --> 00:06:23.918
Meaning when they controlled for alpha momentum in their statistical tests,

235
00:06:24.394 --> 00:06:27.996
the predictive power of simple price momentum just disappeared.

236
00:06:28.082 --> 00:06:29.503
It became insignificant.

237
00:06:29.644 --> 00:06:29.918
Wow.

238
00:06:30.324 --> 00:06:31.082
But crucially,

239
00:06:31.628 --> 00:06:35.878
alpha momentum remained significant even when they controlled for price momentum.

240
00:06:36.128 --> 00:06:38.785
So the risk-adjusted performance is the real driver,

241
00:06:38.847 --> 00:06:40.285
not just the price trend itself.

242
00:06:40.488 --> 00:06:42.113
It strongly suggests that yes,

243
00:06:42.410 --> 00:06:43.707
it's a more fundamental signal,

244
00:06:43.769 --> 00:06:44.113
apparently.

245
00:06:44.438 --> 00:06:44.998
Fascinating.

246
00:06:45.298 --> 00:06:46.479
What about the reversal side?

247
00:06:46.598 --> 00:06:48.278
Alpha reversal versus price reversal.

248
00:06:48.319 --> 00:06:49.057
Was the same story?

249
00:06:49.659 --> 00:06:51.838
It was a bit more complicated there.

250
00:06:52.018 --> 00:06:56.561
Alpha reversal did show explanatory power over price reversal in some situations,

251
00:06:56.983 --> 00:06:59.780
but it wasn't quite as clean cut as the momentum side.

252
00:06:59.999 --> 00:07:02.077
Price reversal sometimes still held significance.

253
00:07:02.389 --> 00:07:02.608
Okay,

254
00:07:02.624 --> 00:07:04.967
so maybe a bit more going on with the long term reversals.

255
00:07:05.389 --> 00:07:08.999
Did they look at why these effects might be stronger in certain markets,

256
00:07:09.045 --> 00:07:10.467
like limits to arbitrage?

257
00:07:10.750 --> 00:07:11.171
They did.

258
00:07:11.331 --> 00:07:14.093
They explored whether things that make arbitrage harder,

259
00:07:14.374 --> 00:07:15.614
like smaller market size,

260
00:07:15.954 --> 00:07:17.478
higher company-specific risk,

261
00:07:17.735 --> 00:07:18.935
idiosyncratic volatility,

262
00:07:19.497 --> 00:07:23.122
or lower correlation with the global market affected the profitability.

263
00:07:23.560 --> 00:07:25.544
And did those factors matter?

264
00:07:25.966 --> 00:07:26.208
Yes,

265
00:07:26.528 --> 00:07:27.161
significantly.

266
00:07:27.489 --> 00:07:32.927
Both the AMO and ARV strategies were much more profitable in markets with higher limits to arbitrage.

267
00:07:33.271 --> 00:07:33.474
Okay,

268
00:07:33.536 --> 00:07:37.536
so where it's harder for big players to create these effects away,

269
00:07:38.130 --> 00:07:39.833
the anomalies persist more strongly.

270
00:07:39.962 --> 00:07:40.423
Exactly.

271
00:07:40.663 --> 00:07:42.085
They showed some clear examples.

272
00:07:42.124 --> 00:07:46.448
The alpha spreads between the long and short portfolios were much wider in,

273
00:07:46.589 --> 00:07:46.788
say,

274
00:07:47.050 --> 00:07:49.890
smaller markets or markets with higher volatility.

275
00:07:50.491 --> 00:07:53.718
It was also stronger in emerging markets compared to developed ones.

276
00:07:53.773 --> 00:07:54.257
Makes sense.

277
00:07:54.874 --> 00:07:57.476
Arbitrage constraints letting these inefficiencies linger.

278
00:07:58.319 --> 00:07:59.741
Did the effects hold up over time?

279
00:07:59.757 --> 00:08:01.179
Did they look at different superiors?

280
00:08:01.180 --> 00:08:01.601
They did.

281
00:08:01.741 --> 00:08:05.163
The momentum strategy seemed pretty stable across different time periods,

282
00:08:05.164 --> 00:08:06.054
which is encouraging.

283
00:08:06.132 --> 00:08:06.304
Good.

284
00:08:06.546 --> 00:08:07.027
Interestingly,

285
00:08:07.067 --> 00:08:11.110
they found the alpha momentum effect was actually stronger following bull markets.

286
00:08:11.332 --> 00:08:11.473
Huh.

287
00:08:12.391 --> 00:08:13.113
Any theories why?

288
00:08:13.352 --> 00:08:18.754
Maybe behavioral biases getting amplified or trends persisting more strongly when sentiment is high.

289
00:08:19.238 --> 00:08:20.863
Hard to say definitively from this.

290
00:08:21.621 --> 00:08:22.488
On the flip side,

291
00:08:22.535 --> 00:08:26.567
the reversal effects seem to get a bit weaker in the later years of their sample.

292
00:08:26.863 --> 00:08:26.988
OK,

293
00:08:27.129 --> 00:08:28.785
so momentum holding up,

294
00:08:28.973 --> 00:08:30.785
reversal perhaps fading slightly.

295
00:08:31.926 --> 00:08:33.238
Did market conditions matter,

296
00:08:33.473 --> 00:08:34.754
like high stress periods?

297
00:08:34.832 --> 00:08:35.098
Yes,

298
00:08:35.238 --> 00:08:35.895
they checked that too.

299
00:08:36.782 --> 00:08:37.803
The reversal effect,

300
00:08:37.983 --> 00:08:38.404
AREV,

301
00:08:38.863 --> 00:08:42.707
was stronger during periods associated with high limits to arbitrage,

302
00:08:42.750 --> 00:08:47.336
like when the VIX was high or credit spreads were wide or the TED spread spiked.

303
00:08:47.836 --> 00:08:50.398
Also during times of high investor sentiment.

304
00:08:50.633 --> 00:08:55.773
So reversal gets amplified when markets are stressed or perhaps overly optimistic.

305
00:08:55.804 --> 00:08:57.023
That's another layer to consider.

306
00:08:57.304 --> 00:08:57.789
Absolutely.

307
00:08:58.117 --> 00:09:00.023
It suggests these aren't just constant effects.

308
00:09:00.289 --> 00:09:03.633
Their strength can ebb and flow with broader market conditions and sentiment.

309
00:09:03.970 --> 00:09:06.971
This has been a really insightful look at alpha momentum and reversal.

310
00:09:07.012 --> 00:09:09.871
It definitely moves beyond simple price trends.

311
00:09:10.012 --> 00:09:10.211
Yeah,

312
00:09:10.332 --> 00:09:16.613
it suggests that looking at that risk-adjusted performance history might offer a more refined edge for global allocation,

313
00:09:16.691 --> 00:09:18.113
both short-term and long-term.

314
00:09:18.433 --> 00:09:20.816
Thank you for tuning in to Papers with Backtest podcast.

315
00:09:21.113 --> 00:09:23.238
We hope today's episode gave you useful insights.

316
00:09:23.660 --> 00:09:25.660
Join us next time as we break down more research.

317
00:09:25.847 --> 00:09:27.316
And for more papers and backtests,

318
00:09:27.332 --> 00:09:30.457
find us at https.paperswithbacktest.com.

319
00:09:31.441 --> 00:09:32.097
Happy trading.