Forecasting and weekly data WIP

Forecasting.ipynb

@@ -0,0 +1,65 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "# Forecasting\n",
+    "\n",
+    "In order to create better portfolios, we need to research best possible forecasting approach, for both the covariance matrix and the expected returns.\n",
+    "\n",
+    "## Covariance\n",
+    "\n",
+    "Baseline: histroical covariance / CovarianceShrinkage taken from PyPortfolioOpt\n",
+    "\n",
+    "References:\n",
+    "- [Summary of classical forecasting methods](https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2411493)\n",
+    "- [Paper doing similar work to ours, with interesing strategies to extract longer-period values from short-period predictions](http://econ.au.dk/fileadmin/site_files/filer_oekonomi/Working_Papers/CREATES/2014/rp14_42.pdf?fbclid=IwAR1PD0w4EFzq6MfJlAlzjDVCZpXmER5sVVjM7tkIVzbogZLitIaGpgGhBY8)\n",
+    "\n",
+    "## Expected returns\n",
+    "\n",
+    "Baseline: mean historical yearly returns\n",
+    "\n",
+    "Notes:\n",
+    "- might be unpredictable\n",
+    "\n",
+    "References: -\n",
+    "\n",
+    "## Dataset construction\n",
+    "Before starting work on these methods, gather as much historical data as possible.\n",
+    "\n",
+    "See `DatasetCreation` notebook for more info."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "#TODO"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.7.2"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}