Fixed notebooks

notebooks/05 Timeseries Forecasting.ipynb

@@ -215,7 +215,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "we now have 1252 unique windows (identified by stock symbol and ending date):"
+    "we now have 1253 unique windows (identified by stock symbol and ending date):"
    ]
   },
   {
@@ -342,7 +342,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "Just to repeat: the features in this row were only calculated using the time series values of `AAPL` up to and including `2015-07-14` and the last 20 days."
+    "Just to repeat: the features in this row were only calculated using the time series values of `AAPL` up to and including `2020-07-14` and the last 20 days."
    ]
   },
   {
@@ -418,7 +418,7 @@
    "source": [
     "We can now train normal AdaBoostRegressors to predict the next time step .\n",
     "Let's split the data into a training and testing sample (but make sure to keep temporal consistency).\n",
-    "We take everything until 2019 as train data an the rest as test:"
+    "We take everything until 2020 as train data an the rest as test:"
    ]
   },
   {
@@ -427,7 +427,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "X[:\"2018\"]"
+    "X[:\"2020\"]"
    ]
   },
   {
@@ -436,11 +436,11 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "X_train = X[:\"2018\"]\n",
-    "X_test = X[\"2019\":]\n",
+    "X_train = X[:\"2020\"]\n",
+    "X_test = X[\"2021\":]\n",
     "\n",
-    "y_train = y[:\"2018\"]\n",
-    "y_test = y[\"2019\":]"
+    "y_train = y[:\"2020\"]\n",
+    "y_test = y[\"2021\":]"
    ]
   },
   {
@@ -535,7 +535,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.10.9"
+   "version": "3.10.13"
   }
  },
  "nbformat": 4,

notebooks/advanced/05 Timeseries Forecasting (multiple ids).ipynb

@@ -197,7 +197,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "Just to repeat: the features in this row were only calculated using the time series values of `AAPL` up to and including `2015-07-15` and the last 20 days."
+    "Just to repeat: the features in this row were only calculated using the time series values of `AAPL` up to and including `2020-07-14` and the last 20 days."
    ]
   },
   {
@@ -268,11 +268,11 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "X_train = X.loc[(slice(None), slice(None, \"2018\")), :]\n",
-    "X_test = X.loc[(slice(None), slice(\"2019\", \"2020\")), :]\n",
+    "X_train = X.loc[(slice(None), slice(None, \"2020\")), :]\n",
+    "X_test = X.loc[(slice(None), slice(\"2021\", \"2022\")), :]\n",
     "\n",
-    "y_train = y.loc[(slice(None), slice(None, \"2018\"))]\n",
-    "y_test = y.loc[(slice(None), slice(\"2019\", \"2020\"))]"
+    "y_train = y.loc[(slice(None), slice(None, \"2020\"))]\n",
+    "y_test = y.loc[(slice(None), slice(\"2021\", \"2022\"))]"
    ]
   },
   {
@@ -336,6 +336,13 @@
     "y.unstack(\"Symbols\").plot(ax=plt.gca())\n",
     "y_pred.unstack(\"Symbols\").plot(ax=plt.gca(), legend=None, marker=\".\")"
    ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
   }
  ],
  "metadata": {
@@ -354,7 +361,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.7.7"
+   "version": "3.10.13"
   }
  },
  "nbformat": 4,

notebooks/advanced/visualize-benjamini-yekutieli-procedure.ipynb

@@ -38,8 +38,7 @@
     "from tsfresh.feature_extraction import ComprehensiveFCParameters\n",
     "\n",
     "matplotlib.rcParams[\"figure.figsize\"] = [16, 6]\n",
-    "matplotlib.rcParams[\"font.size\"] = 14\n",
-    "matplotlib.style.use('seaborn-darkgrid')"
+    "matplotlib.rcParams[\"font.size\"] = 14"
    ]
   },
   {