fix: fill WRMSSE levels 1-5, regenerate parity fixtures (5k rows), recompute SHA256SUMS

SHA256SUMS

@@ -1,10 +1,21 @@
 1e3e0d86934828b0a062f6162eb59ddc9a021fb719893e351ff38fe328ddac29  M5_EXPLAINED.md
 3ca7537ecf7a5f205e942b8c3ddb3226c71e2f509b32fc77e2ef29a10d4b0176  README.md
-a192f23a1917dda6870b8365d319faa0166a53b20518bb982edf073eec7687a0  SHA256SUMS
+4f0539c4603e352381006175c68b443220e73e83144e9a58b712d0443f15e8e2  SHA256SUMS
 9dc75d44cb5885843e1f625d1df47b7d9d1ce988a49977b1c1bce4d936cabf34  TECHNICAL_REPORT.md
+df67405ffc715a2d5fefbf39d327b05bc1ddcf35fcdcab03352e116e0f4fc967  arima/README.md
+b491f07563188513c32e25f0ddc345055522106d0de3a49beb4d9fc313fc38ed  arima/metrics.json
+9a186737a57e95e4769377efca1d3859683eff2a2cde59ff52c88028fc88f188  arima/model.pkl
+ef416fd5e1737527027f724b71a81904b84145fc95fe558b945d3b15577a46d4  arima/model_config.json
+1d445ed6e209060921db3b045ff69737977340b9f7e1dfaeca8a7bddb203b84c  arima/predictions.csv
+35dc02095bc79aadf79b8ccce25c126dcf200edc35a20ce4a0bf8c570583bba9  arima_phased/model.pkl
+57b2f694c181010dc63b01c1c93108df22afab75a84b639042e09ae8f1cb7894  arima_phased/model_config.json
 550779bc6804a34668d6db43dd7b57df23a3bc465b456874437d03940a77439b  categorical_maps.json
 fa03349beb59db7282e6557c6d24f74fcfe6429537fcbeb024fd041bbe16b633  feature_schema.json
 a4a50978a69487c1e8af483e046492a768decb2e942fa34e3887ed1cccd50de3  history_tail.parquet
+f1a73c66a24b081be927b25cd1063eeb8cb5ab96031b1f9d186c01bc41940af1  hybrid/model_config.json
+7390b063360318d11662174be4cdc13f4b993c0bb56790432c13d9b087ce6f9f  hybrid/predictions.csv
+782b507b7ee51efa50e6b1177b4852cbfe92cc05ac461fd420497fcc55a0831e  hybrid/sarimax_base.pkl
+611f01ff33b1caae26f52de56db27f167981c2e1fa5669d005d45967af9abe28  hybrid/xgboost_residual.pkl
 61420adbe7f965b5b3e83f99ab911fd59d0b134400c82714e809feabcf88f4e6  models/model_store=CA_1_hblock=block_15_21.txt
 9e3eeb78f17418a7cf3ad9f4a3f5856abaac3878cf65250ec3077df0576d6b2e  models/model_store=CA_1_hblock=block_1_7.txt
 71a9c3bd31668caa35f3c53b15a9e9a7b0897fd4317430cfb391bb0a08e5972a  models/model_store=CA_1_hblock=block_22_28.txt
@@ -45,8 +56,23 @@ e7c4f51979fc293a024968c9c16734bbad193b2410424262396d1b62927d7fa5  models/model_s
 1ba168cd8b190468c099292da4f8d2fae4d7cfbc17d01319c3211e05e093eceb  models/model_store=WI_3_hblock=block_1_7.txt
 2c573b52d611ffce903272a9b0c394bed00e8d092f70dff1b4d4ea59a8c19647  models/model_store=WI_3_hblock=block_22_28.txt
 179fac6439cdba03422554b551e7361745396372b2e6a96af2fa70a5807eda31  models/model_store=WI_3_hblock=block_8_14.txt
-432abac6eb8e984979406701d8d6169202739941722b9c804c2f9df960ea22a6  parity_predictions.parquet
-dddf3c5adbad6ebe33e004fbc34d5c78dab974c0dcfbc5400b98d0774681e620  predictions.parquet
+770f6433d5104a024dbb067cf7d7f101987a8f6bc47f65b8e701638c0f36f78c  parity_fixture.parquet
+5f0b261b519c608137ffbf23ef312008ec14405804e750a575dfd6c464e7f70b  parity_predictions.parquet
+20e8563186f83d8e20836d6959f68f6870aada40e60e07ac6eef50d8b13ec327  predictions.parquet
 76a7801755bbb5d52829d5e97bbad88decba57a9e48cc8a73d7911dd0a298f66  predictions_meta.json
-b953f9c0d694aed270eed0e883501aa4026c6e216f3fa97c14727f82895922c6  submission.csv
-ceaa3584d31e7ca341ac5d641eaae981e0f8446a2925cc9e842e40db9dbb5026  training_config.json
+0c33a06ceec528ae24d33e2ddb3f5edaf37731a31212448228602001c9163d4c  prophet/README.md
+562c41fb80ebdbf2c7c274349c44ce2bf9351462659a5f77b7a645506063d338  prophet/metrics.json
+df52719ab0b801c14080bd001f041825bd7271fc867fbacae93442f396909b7a  prophet/model.pkl
+1882907403b41cad5b6e029b006d1a0e8f0f6358eefe9a41459024c46d7c2695  prophet/model_config.json
+142a0d04f6eeeaf0594240bafa7b195210ba457e52eb4ae8fe15031350e83f01  prophet/predictions.csv
+b800ee71f028a519babe82758de8283e9c332a13988008fb3c2093f079097923  prophet_phased/model.pkl
+466cd0eb6efc9947fe7f0465642e955ab725190f2d35312108080aa473bd8c53  prophet_phased/model_config.json
+efb8ae29223b777e0f79159bda13e18375cc0cfe9908afc36b3078e61b7b9687  sarimax/README.md
+75bafa27684334005f586c86cabf54c3b653e1b22576f8da27c6f1a2520299b4  sarimax/metrics.json
+97c55e424c49326ad5989df42aab3a398158b8fbaa730e0a465ebfcc111c8ffe  sarimax/model.pkl
+1a0d7d1ed4967accec319cbbef0c4b90c919247038690bc9bf2ba7f3cf78c4fc  sarimax/model_config.json
+41f5e8272b08a0e9c7cd8db3e6f89489b422c27ed2d55eec4384803081746a43  sarimax/predictions.csv
+e689edccdd92359bc62791e29fc78eadadb19673cdc47d98e3bbe45910c8cd10  sarimax_phased/model.pkl
+946236bdbbfe6aa33162fda73f8424d83bb72c0ee5c55064a49bff4262198d8a  sarimax_phased/model_config.json
+2c1e295466a8d2f452070e9b37a8640e1817c41324fc7af11315f5a48bd265ad  submission.csv
+46971a0a8e8fe5fdb0842f67f51f607b323895a1928c48a6ef9cf4bf4f56a6f8  training_config.json

arima/README.md

@@ -0,0 +1,45 @@
+# ARIMA Model for M5 Demand Forecasting
+
+## Overview
+AutoRegressive Integrated Moving Average trained on aggregated daily M5 sales data.
+
+## Model Details
+- **Architecture**: ARIMA(3,1,1) with trend
+- **Training Data**: 1,913 days of aggregated daily sales
+- **Test Period**: 28 days (2016-04-25 to 2016-05-22)
+- **Differencing**: d=1 (first differencing required for stationarity)
+
+## Stationarity Test
+- ADF Statistic: -1.565373
+- p-value: 0.500960
+- Original data non-stationary; first differencing achieves stationarity
+
+## Performance
+| Metric | Value |
+|--------|-------|
+| RMSE | 6,459.70 |
+| MAE | 4,852.62 |
+| MAPE | 10.48% |
+| AIC | 37,801.28 |
+
+## Key Features
+- AR component: 3 autoregressive terms
+- I component: 1st order differencing (d=1)
+- MA component: 1 moving average term
+- Trend term: 't' (deterministic trend)
+
+## Usage
+```python
+import pickle
+
+with open('model.pkl', 'rb') as f:
+    model = pickle.load(f)
+
+forecast = model.forecast(steps=28)
+```
+
+## Notes
+- Simplest model but requires manual order selection
+- No seasonal component (SARIMAX handles this better)
+- Good baseline but underperforms compared to Prophet and SARIMAX
+- Smallest model file (6.5MB) due to no seasonal components

arima/metrics.json

@@ -0,0 +1,15 @@
+{
+  "rmse": 6459.701299981527,
+  "mae": 4852.6228457663765,
+  "mape": 10.475306061903403,
+  "method": "ARIMA",
+  "order": [
+    3,
+    1,
+    1
+  ],
+  "d_value": 1,
+  "train_days": 1913,
+  "test_days": 28,
+  "adf_pvalue": 0.5009604361797737
+}

arima/model.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9a186737a57e95e4769377efca1d3859683eff2a2cde59ff52c88028fc88f188
+size 6870504

arima/model_config.json

@@ -0,0 +1,36 @@
+{
+  "model_name": "ARIMA",
+  "hyperparameters": {
+    "order": [
+      3,
+      1,
+      1
+    ],
+    "trend": "t",
+    "stationary": false
+  },
+  "training_data": {
+    "train_days": 1913,
+    "test_days": 28,
+    "train_start": "2011-01-29 00:00:00",
+    "train_end": "2016-04-24 00:00:00",
+    "stationarity_test": "ADF",
+    "adf_statistic": -1.5653733253318483,
+    "adf_pvalue": 0.5009604361797737
+  },
+  "metrics": {
+    "rmse": 6459.701299981527,
+    "mae": 4852.6228457663765,
+    "mape": 10.475306061903403,
+    "method": "ARIMA",
+    "order": [
+      3,
+      1,
+      1
+    ],
+    "d_value": 1,
+    "train_days": 1913,
+    "test_days": 28,
+    "adf_pvalue": 0.5009604361797737
+  }
+}

arima/predictions.csv

@@ -0,0 +1,29 @@
+ds,yhat,actual
+2016-04-25,45190.990259355225,38793
+2016-04-26,40583.77766973429,35487
+2016-04-27,39073.7103863895,34445
+2016-04-28,40454.294880894406,34732
+2016-04-29,42624.609787427515,42896
+2016-04-30,43799.831241559645,50429
+2016-05-01,43589.1120619281,53032
+2016-05-02,42700.7718190461,43181
+2016-05-03,42017.34031237138,44314
+2016-05-04,41926.30052133391,39601
+2016-05-05,42256.61094459382,40763
+2016-05-06,42620.14437307269,43805
+2016-05-07,42762.768288819745,54239
+2016-05-08,42682.846685461045,45609
+2016-05-09,42530.503889281834,46400
+2016-05-10,42444.48207322392,39379
+2016-05-11,42462.53115942328,42248
+2016-05-12,42536.51388263449,40503
+2016-05-13,42599.7324861895,44073
+2016-05-14,42619.7006711,54308
+2016-05-15,42606.40123128623,59921
+2016-05-16,42588.37788685298,42362
+2016-05-17,42585.86841938702,38777
+2016-05-18,42600.41418280065,37096
+2016-05-19,42621.35490978869,36963
+2016-05-20,42637.98826800487,42552
+2016-05-21,42646.85333683171,51518
+2016-05-22,42651.15957858302,54338

arima_phased/model.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:35dc02095bc79aadf79b8ccce25c126dcf200edc35a20ce4a0bf8c570583bba9
+size 6619306

arima_phased/model_config.json

@@ -0,0 +1,22 @@
+{
+  "model": "ARIMA",
+  "order": [
+    2,
+    1,
+    2
+  ],
+  "d_value": 1,
+  "adf_pvalue": 0.48920532220127466,
+  "test_order": [
+    2,
+    1,
+    1
+  ],
+  "metrics": {
+    "name": "ARIMA",
+    "rmse": 6558.0098488097265,
+    "mae": 5298.588749100937,
+    "mape": 12.038072720026841
+  },
+  "aic": 37119.59765269647
+}

hybrid/model_config.json

@@ -0,0 +1,33 @@
+{
+  "model_name": "Hybrid_SARIMAX_XGBoost",
+  "base_model": "SARIMAX(2,1,1)(1,1,1,7)",
+  "residual_model": "XGBoost",
+  "strategy": "SARIMAX + XGBoost(residual_correction)",
+  "sarimax_rmse": 1805.0597898619735,
+  "hybrid_rmse": 1756.9378292743959,
+  "improvement_pct": 2.665948289239624,
+  "metrics": {
+    "name": "Hybrid_SARIMAX_XGBoost",
+    "rmse": 1756.9378292743959,
+    "mae": 1525.9386274679296,
+    "mape": 3.73186543749005
+  },
+  "features": [
+    "residual_lag_1",
+    "residual_lag_2",
+    "residual_lag_7",
+    "residual_lag_14",
+    "residual_lag_28",
+    "rolling_mean_7",
+    "rolling_mean_28",
+    "rolling_std_7",
+    "sales_lag_1",
+    "sales_lag_7",
+    "wday",
+    "month",
+    "snap_CA",
+    "snap_TX",
+    "snap_WI",
+    "sarimax_fitted"
+  ]
+}

hybrid/predictions.csv

@@ -0,0 +1,29 @@
+date,sarimax_pred,xgboost_correction,hybrid_pred,actual
+2016-03-28,38235.134864620864,-303.56647,37931.56839733571,36041
+2016-03-29,35938.24562629463,-595.3925,35342.853109204785,33857
+2016-03-30,36067.09610403734,-484.44943,35582.64667166429,32359
+2016-03-31,36632.14476851042,-605.6719,36026.47289351042,34681
+2016-04-01,43613.110228852085,501.40698,44114.51721127396,45536
+2016-04-02,53429.26630673088,302.20306,53731.469370695726,52672
+2016-04-03,53326.93428310654,550.79254,53877.72682461044,56425
+2016-04-04,40897.55973830253,-381.6515,40515.90824904472,40418
+2016-04-05,40665.49159821007,-40.607292,40624.88430603478,39683
+2016-04-06,40821.239280567286,776.40137,41597.640647754786,39134
+2016-04-07,40374.93390058706,218.52983,40593.463731519674,38116
+2016-04-08,42826.07274464186,1401.3386,44227.411367688735,43220
+2016-04-09,54621.33167223622,532.9614,55154.29309801747,56340
+2016-04-10,51202.217755138554,-342.1888,50860.028942882695,53856
+2016-04-11,41978.6745867846,487.2263,42465.90087462639,42427
+2016-04-12,39035.19301750588,847.0511,39882.24410393166,39069
+2016-04-13,38016.02303935446,391.74045,38407.76348735251,35193
+2016-04-14,38490.77302491779,549.0712,39039.844252945135,37529
+2016-04-15,42545.122587171194,554.77106,43099.8936443001,41789
+2016-04-16,50277.31445115759,-663.08167,49614.232786118526,48362
+2016-04-17,49596.870777660835,849.41815,50446.288929516304,51640
+2016-04-18,39557.57946680763,-914.13855,38643.44091700294,38059
+2016-04-19,36604.45188763487,-584.38947,36020.062422302835,37570
+2016-04-20,36764.46961673701,-726.60034,36037.86927494014,35343
+2016-04-21,37521.78089081055,-739.74634,36782.034552919926,35033
+2016-04-22,40100.54939246655,-825.13635,39275.41303992749,40517
+2016-04-23,50473.722035745894,-942.18225,49531.53978476933,48962
+2016-04-24,49806.76286619581,-2870.2925,46936.47038572706,49795

hybrid/sarimax_base.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:782b507b7ee51efa50e6b1177b4852cbfe92cc05ac461fd420497fcc55a0831e
+size 88221776

hybrid/xgboost_residual.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:611f01ff33b1caae26f52de56db27f167981c2e1fa5669d005d45967af9abe28
+size 123328