{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import panel as pn\n",
    "import pandas as pd\n",
    "import scipy.stats as stats\n",
    "from sqlalchemy import create_engine\n",
    "import hvplot.pandas\n",
    "pn.extension()\n",
    "db_url = 'sqlite:///instance/local.db'\n",
    "engine = create_engine(db_url)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Expirement to use z-score to find outlier of daily pct, not sure if this is helpful but here is it"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "b_stocks_df = None\n",
    "# load benchmark stock\n",
    "with engine.connect() as connection:\n",
    "    b_stocks_df = pd.read_sql('calculated_b_stock', con=connection)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### z = (x - μ) / σ"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "b_stocks_df['pct_mean'] = b_stocks_df.groupby('ticker')['pct']\\\n",
    "    .transform(lambda x: x.rolling(7, min_periods=1).mean())\n",
    "b_stocks_df['pct_std'] = b_stocks_df.groupby('ticker')['pct']\\\n",
    "    .transform(lambda x: x.rolling(7, min_periods=1).std())\n",
    "b_stocks_df['pct_z_score'] = (b_stocks_df['pct'] - b_stocks_df['pct_mean']) / b_stocks_df['pct_std']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "select = pn.widgets.Select(name='Select', options=b_stocks_df.display_name.unique().tolist())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "ib_stocks_df = b_stocks_df.interactive()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "selected_stock_df = ib_stocks_df[ib_stocks_df.display_name == select]\n",
    "outliers = selected_stock_df[((ib_stocks_df['pct_z_score'] > 2) | (ib_stocks_df['pct_z_score'] < -2))]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "outlier_plot = outliers.hvplot.scatter(x='date',y='pct',color='red')\n",
    "pct_plot = selected_stock_df.hvplot(x='date',y='pct')\n",
    "outlier_plot * pct_plot\n"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "portfolio_risk_assesment",
   "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": "3.11.4"
  },
  "orig_nbformat": 4
 },
 "nbformat": 4,
 "nbformat_minor": 2
}