Repository navigation
Expand file tree
/
Copy pathfeed.xml
More file actions
234 lines (234 loc) · 17.2 KB
/
Copy pathfeed.xml
File metadata and controls
234 lines (234 loc) · 17.2 KB
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
<?xml version="1.0" encoding="utf-8"?>
<feed xmlns="http://www.w3.org/2005/Atom">
<title>QuantLab Research Notes</title>
<link rel="alternate" href="https://alencheung.github.io/quantlab-notes/"/>
<link rel="self" href="https://alencheung.github.io/quantlab-notes/feed.xml"/>
<id>https://alencheung.github.io/quantlab-notes/feed.xml</id>
<updated>2026-09-01T00:00:00Z</updated>
<author><name>QuantLab Research Desk</name></author>
<entry>
<title>1 hit, 0 misses: the Track Record's first graded call</title>
<link rel="alternate" href="https://alencheung.github.io/quantlab-notes/articles/track-record-first-hit/"/>
<id>https://alencheung.github.io/quantlab-notes/articles/track-record-first-hit/</id>
<published>2026-09-01T00:00:00Z</published>
<updated>2026-09-01T00:00:00Z</updated>
<summary>The ledger's first maturity arrived this cycle and the grader stamped it: the $75,000 BTC conviction call is a hit - cleared on session 1 of 10 at $77,525 and never closed below it - the Track Record reads 1 hit, 0 misses, 6 open.</summary>
<author><name>QuantLab Research Desk</name></author>
</entry>
<entry>
<title>+0.31 into the drawdown: bonds still refuse to hedge</title>
<link rel="alternate" href="https://alencheung.github.io/quantlab-notes/articles/rates-corr-september-state/"/>
<id>https://alencheung.github.io/quantlab-notes/articles/rates-corr-september-state/</id>
<published>2026-09-01T00:00:00Z</published>
<updated>2026-09-01T00:00:00Z</updated>
<summary>The 60-day stock-bond correlation printed its call-window minimum on Tuesday (+0.31) as the streak reached 164 sessions - and the five-session equity dip put the non-hedge on tape: SPY -0.54% with TLT -1.92% beside it.</summary>
<author><name>QuantLab Research Desk</name></author>
</entry>
<entry>
<title>9.5 points top to bottom: August's sector scoreboard</title>
<link rel="alternate" href="https://alencheung.github.io/quantlab-notes/articles/sector-lab-august-scoreboard/"/>
<id>https://alencheung.github.io/quantlab-notes/articles/sector-lab-august-scoreboard/</id>
<published>2026-08-30T00:00:00Z</published>
<updated>2026-08-30T00:00:00Z</updated>
<summary>August through Friday splits the nine sector SPDRs by 9.5 points - tech +5.9% and the materials/health/energy cluster just behind, utilities -3.7% dead last and alone below its 200-day mean - while energy still runs +42.1% year-to-date.</summary>
<author><name>QuantLab Research Desk</name></author>
</entry>
<entry>
<title>P(risk) 0.0004: the regime model's verdict on the pullback</title>
<link rel="alternate" href="https://alencheung.github.io/quantlab-notes/articles/regime-radar-002/"/>
<id>https://alencheung.github.io/quantlab-notes/articles/regime-radar-002/</id>
<published>2026-08-29T00:00:00Z</published>
<updated>2026-08-29T00:00:00Z</updated>
<summary>Seven sessions after issue 1, SPY has given back 1.10% and the live book took its first drawdown - and the causal HMM's risk posterior FELL from 0.0006 to 0.0004, staying in the bull state: the radar's verdict is that this is not a regime event.</summary>
<author><name>QuantLab Research Desk</name></author>
</entry>
<entry>
<title>-1.03%: the live book's first drawdown</title>
<link rel="alternate" href="https://alencheung.github.io/quantlab-notes/articles/live-desk-first-drawdown/"/>
<id>https://alencheung.github.io/quantlab-notes/articles/live-desk-first-drawdown/</id>
<published>2026-08-28T00:00:00Z</published>
<updated>2026-08-28T00:00:00Z</updated>
<summary>Six sessions in, the paper-live book just took its first drawdown - three straight red days totaling -1.03%, the 7.1st percentile of its own backtest's three-day noise, with 96% of the distance to the first kill-switch latch still in reserve.</summary>
<author><name>QuantLab Research Desk</name></author>
</entry>
<entry>
<title>2.46x median and drifting down: the liquidity call's long road back</title>
<link rel="alternate" href="https://alencheung.github.io/quantlab-notes/articles/liquidity-dv63-drift/"/>
<id>https://alencheung.github.io/quantlab-notes/articles/liquidity-dv63-drift/</id>
<published>2026-08-27T00:00:00Z</published>
<updated>2026-08-27T00:00:00Z</updated>
<summary>Our standing call says 63-day dollar volume falls back under 2x its long-run median; it has printed 2.46x through Wednesday - and the decomposition says the call is secretly a price call, because share volume is 0.80x its own median while prices run 2.77x theirs.</summary>
<author><name>QuantLab Research Desk</name></author>
</entry>
<entry>
<title>8 for 8 and 5 for 5: the squeeze settled both BTC calls</title>
<link rel="alternate" href="https://alencheung.github.io/quantlab-notes/articles/btc-squeeze-settles-calls/"/>
<id>https://alencheung.github.io/quantlab-notes/articles/btc-squeeze-settles-calls/</id>
<published>2026-08-27T00:00:00Z</published>
<updated>2026-08-27T00:00:00Z</updated>
<summary>Both standing BTC calls look resolved: every one of the 8 window sessions since the reclaim call's anchor closed above the 200-day mean, all 5 since the $75,000 call's anchor closed above $75,000, and BTC now prints $80,264 - 27.8% off the August 1 low.</summary>
<author><name>QuantLab Research Desk</name></author>
</entry>
<entry>
<title>118 of 159 days above 2.0, then 1.47: the VRP compression the call is waiting out</title>
<link rel="alternate" href="https://alencheung.github.io/quantlab-notes/articles/vrp-compression-watch/"/>
<id>https://alencheung.github.io/quantlab-notes/articles/vrp-compression-watch/</id>
<published>2026-08-26T00:00:00Z</published>
<updated>2026-08-26T00:00:00Z</updated>
<summary>The standing call says the VRP recovers above 2.0 vol points within 60 sessions; the grader last saw 1.47 on August 20, while Cboe's fresher spot print implies an estimated 2.16 - and the call now waits on a cache that refreshes in September.</summary>
<author><name>QuantLab Research Desk</name></author>
</entry>
<entry>
<title>9 sessions, 1.52% short: the record-reclaim clock</title>
<link rel="alternate" href="https://alencheung.github.io/quantlab-notes/articles/spy-ath-reclaim-clock/"/>
<id>https://alencheung.github.io/quantlab-notes/articles/spy-ath-reclaim-clock/</id>
<published>2026-08-26T00:00:00Z</published>
<updated>2026-08-26T00:00:00Z</updated>
<summary>Our standing call says SPY reclaims its August 13 record within 20 sessions; the spell is 9 sessions old and 1.52% short with 16 sessions left - and every one of the 14 completed shallow spells that ever reached age 9 was resolved by age 25.</summary>
<author><name>QuantLab Research Desk</name></author>
</entry>
<entry>
<title>5 of 6 pass, 0 admitted: wing rotation survives costs but not the diversity rule</title>
<link rel="alternate" href="https://alencheung.github.io/quantlab-notes/articles/wing-rotation-six-cells/"/>
<id>https://alencheung.github.io/quantlab-notes/articles/wing-rotation-six-cells/</id>
<published>2026-08-25T00:00:00Z</published>
<updated>2026-08-25T00:00:00Z</updated>
<summary>We pre-registered six strategies that rotate between the library's momentum and defensive wings; our first run mis-charged their trading costs 10x, and at the correct 3.5 bps five of six pass the battery - but all five are refused by the corr-below-0.75 diversity rule at 0.77-0.94.</summary>
<author><name>QuantLab Research Desk</name></author>
</entry>
<entry>
<title>11 of 16 for three straight sessions: breadth knocks on the 70% door</title>
<link rel="alternate" href="https://alencheung.github.io/quantlab-notes/articles/rotation-breadth-one-tick/"/>
<id>https://alencheung.github.io/quantlab-notes/articles/rotation-breadth-one-tick/</id>
<published>2026-08-25T00:00:00Z</published>
<updated>2026-08-25T00:00:00Z</updated>
<summary>Our standing call says panel breadth closes above 70% within 20 sessions - it has now printed exactly 11 of 16 ETFs above their 200-day mean for three straight sessions, one tick short, with the entire below-line complex being bonds and utilities.</summary>
<author><name>QuantLab Research Desk</name></author>
</entry>
<entry>
<title>+0.33 and holding: the stock-bond correlation that forgot to break</title>
<link rel="alternate" href="https://alencheung.github.io/quantlab-notes/articles/stock-bond-corr-60d/"/>
<id>https://alencheung.github.io/quantlab-notes/articles/stock-bond-corr-60d/</id>
<published>2026-08-24T00:00:00Z</published>
<updated>2026-08-24T00:00:00Z</updated>
<summary>Our standing call said the 60-day SPY-TLT correlation would break below zero within 166 sessions - through Friday it has spent 157 straight sessions above it, and at +0.33 it now exceeds even its 2022 inflation-shock peak.</summary>
<author><name>QuantLab Research Desk</name></author>
</entry>
<entry>
<title>+0.20% in two sessions: the live ledger opens, and the evidence clock starts</title>
<link rel="alternate" href="https://alencheung.github.io/quantlab-notes/articles/live-desk-first-two-sessions/"/>
<id>https://alencheung.github.io/quantlab-notes/articles/live-desk-first-two-sessions/</id>
<published>2026-08-24T00:00:00Z</published>
<updated>2026-08-24T00:00:00Z</updated>
<summary>Our paper-live book wrote its second hash-chained session and flipped positive (+0.20% cumulative, chain valid, kill stage GREEN) - and the platform's own tested Minimum-Track-Record-Length maths says exactly how much evidence the book still owes: 585 sessions.</summary>
<author><name>QuantLab Research Desk</name></author>
</entry>
<entry>
<title>17 of 20 up: the strategy library's August scoreboard</title>
<link rel="alternate" href="https://alencheung.github.io/quantlab-notes/articles/library-august-scoreboard/"/>
<id>https://alencheung.github.io/quantlab-notes/articles/library-august-scoreboard/</id>
<published>2026-08-23T00:00:00Z</published>
<updated>2026-08-23T00:00:00Z</updated>
<summary>We re-ran all 20 equity strategies through the engine on data through Friday: 17 are up in August with a +1.6% median, the momentum wing leads, and the honest headline is that the defensive book trails a tape that set records into mid-August.</summary>
<author><name>QuantLab Research Desk</name></author>
</entry>
<entry>
<title>55% of weekday vol: the quiet weekend regime on the 58-pair crypto book</title>
<link rel="alternate" href="https://alencheung.github.io/quantlab-notes/articles/crypto-weekends-58-pairs/"/>
<id>https://alencheung.github.io/quantlab-notes/articles/crypto-weekends-58-pairs/</id>
<published>2026-08-23T00:00:00Z</published>
<updated>2026-08-23T00:00:00Z</updated>
<summary>Crypto never closes, but it does doze: on our 58-pair OKX book every pair with a year of history runs quieter on weekends, and BTC's weekend volatility has run at roughly half its weekday level over the last 90 days.</summary>
<author><name>QuantLab Research Desk</name></author>
</entry>
<entry>
<title>Nine sectors, 0.85: the sector book's best trade is the trade we already own</title>
<link rel="alternate" href="https://alencheung.github.io/quantlab-notes/articles/nine-sectors-one-trade/"/>
<id>https://alencheung.github.io/quantlab-notes/articles/nine-sectors-one-trade/</id>
<published>2026-08-22T00:00:00Z</published>
<updated>2026-08-22T00:00:00Z</updated>
<summary>We searched a pure US sector ETF cross-section for the first time - 21 pre-registered constructions, five passed every performance metric, and every one of them correlated 0.77-0.85 with the defensive rotation our 16-ETF library already holds.</summary>
<author><name>QuantLab Research Desk</name></author>
</entry>
<entry>
<title>+21.1% in Three UTC Days: Bitcoin Broke Its Own Rule While SPY Slipped</title>
<link rel="alternate" href="https://alencheung.github.io/quantlab-notes/articles/btc-vs-first-spy-dip/"/>
<id>https://alencheung.github.io/quantlab-notes/articles/btc-vs-first-spy-dip/</id>
<published>2026-08-21T00:00:00Z</published>
<updated>2026-08-21T00:00:00Z</updated>
<summary>BTC compounded +21.1% across the three complete UTC days to Aug 21 while SPY slipped −0.23%, closing $78,409 and +13.6% above its 200-day mean — a break from a 2026 regime of −0.97% SPY-down-day means and 1.62 down-beta.</summary>
<author><name>QuantLab Research Desk</name></author>
</entry>
<entry>
<title>Six Years Under Water: the Bond Spell Nobody Prices and the Anatomy of Recovery</title>
<link rel="alternate" href="https://alencheung.github.io/quantlab-notes/articles/time-under-water/"/>
<id>https://alencheung.github.io/quantlab-notes/articles/time-under-water/</id>
<published>2026-08-20T00:00:00Z</published>
<updated>2026-08-20T00:00:00Z</updated>
<summary>AGG, IEF and TLT have been below their shared 2020 peak for 6.04 years (TLT −41.6%); no asset closed within 0.1% of its high on Aug 19; SPY's median deep-spell recovery is 204 days — duration, not depth, is the risk nobody prices.</summary>
<author><name>QuantLab Research Desk</name></author>
</entry>
<entry>
<title>The $99 Billion Mirage: Record Volume That Is Mostly Record Price</title>
<link rel="alternate" href="https://alencheung.github.io/quantlab-notes/articles/volume-behind-the-record/"/>
<id>https://alencheung.github.io/quantlab-notes/articles/volume-behind-the-record/</id>
<published>2026-08-19T00:00:00Z</published>
<updated>2026-08-19T00:00:00Z</updated>
<summary>Universe dollar volume runs $98.9B a day — triple the median year, #1 of 19 — but share volume is up just 3.4%: the record tape is mostly record price, with liquidity rotating from SPY (−8.6pp) to QQQ (+10.0pp).</summary>
<author><name>QuantLab Research Desk</name></author>
</entry>
<entry>
<title>Who Pays for the Calm? A Variance Risk Premium of 1.3 Points</title>
<link rel="alternate" href="https://alencheung.github.io/quantlab-notes/articles/vol-vrp-anatomy/"/>
<id>https://alencheung.github.io/quantlab-notes/articles/vol-vrp-anatomy/</id>
<published>2026-08-19T00:00:00Z</published>
<updated>2026-08-19T00:00:00Z</updated>
<summary>Implied vol trades just 1.27 points over realized — the 23rd percentile since 2007 — and the forward record says thin VRP is no sell signal either: the vol-carry trade's economics, computed.</summary>
<author><name>QuantLab Research Desk</name></author>
</entry>
<entry>
<title>220 Days of Broken Hedge: the Stock-Bond Correlation Stays Positive</title>
<link rel="alternate" href="https://alencheung.github.io/quantlab-notes/articles/stock-bond-correlation/"/>
<id>https://alencheung.github.io/quantlab-notes/articles/stock-bond-correlation/</id>
<published>2026-08-19T00:00:00Z</published>
<updated>2026-08-19T00:00:00Z</updated>
<summary>The 60-day SPY–TLT correlation has been positive since January 7 — nearly four times the average break — and the risk reduction a 60/40 earns from bonds has halved, from 44% to 22%.</summary>
<author><name>QuantLab Research Desk</name></author>
</entry>
<entry>
<title>Zero Fear at Record Highs — and the Hedge That Stopped Working</title>
<link rel="alternate" href="https://alencheung.github.io/quantlab-notes/articles/regime-radar-001/"/>
<id>https://alencheung.github.io/quantlab-notes/articles/regime-radar-001/</id>
<published>2026-08-19T00:00:00Z</published>
<updated>2026-08-19T00:00:00Z</updated>
<summary>P(bear) = 0.06%, the fear gauge pinned at its 3rd percentile, stocks and bonds now moving together — the August 2026 tape computed from six data sources, with every number and every blind spot disclosed.</summary>
<author><name>QuantLab Research Desk</name></author>
</entry>
<entry>
<title>Twenty Strategies, Two Bets: the Diversification Math Nobody Shows You</title>
<link rel="alternate" href="https://alencheung.github.io/quantlab-notes/articles/library-three-bets/"/>
<id>https://alencheung.github.io/quantlab-notes/articles/library-three-bets/</id>
<published>2026-08-19T00:00:00Z</published>
<updated>2026-08-19T00:00:00Z</updated>
<summary>The certified 20-strategy book earns Sharpe 1.09 by averaging streams that correlate 0.50 on average — an effective N of 1.9 since 2007, and just 1.4 today. Twenty lines on the factsheet; roughly two independent bets in the risk engine.</summary>
<author><name>QuantLab Research Desk</name></author>
</entry>
<entry>
<title>Crypto Doesn't Sleep. It Takes a Lunch Break.</title>
<link rel="alternate" href="https://alencheung.github.io/quantlab-notes/articles/crypto-hourly-structure/"/>
<id>https://alencheung.github.io/quantlab-notes/articles/crypto-hourly-structure/</id>
<published>2026-08-19T00:00:00Z</published>
<updated>2026-08-19T00:00:00Z</updated>
<summary>66,889 hourly bars per symbol: volatility peaks at 14:00 UTC (10 a.m. ET, the heart of the US session), bottoms at 05:00 UTC, weekends run 24% calmer than weekdays — and all eight majors post negative 2026 returns. One beta, eight wrappers.</summary>
<author><name>QuantLab Research Desk</name></author>
</entry>
<entry>
<title>The Narrowest Record in Years: 62.5% Breadth at an All-Time High</title>
<link rel="alternate" href="https://alencheung.github.io/quantlab-notes/articles/breadth-at-records/"/>
<id>https://alencheung.github.io/quantlab-notes/articles/breadth-at-records/</id>
<published>2026-08-19T00:00:00Z</published>
<updated>2026-08-19T00:00:00Z</updated>
<summary>On the 601 days SPY closed at a record since 2007, median panel breadth was 87.5% and 73% of those days were above 80%. Today's record-adjacent breadth is 62.5% — the 13th percentile of record days. The index is narrow; the tape underneath is narrower.</summary>
<author><name>QuantLab Research Desk</name></author>
</entry>
</feed>