Upload arima model with evaluation metrics

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

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
+}

model.pkl

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

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
+  }
+}

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