{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import exploration.marketing as mk\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Vary weights:size of investment and success_prob: likelihood of investment"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "##----Now look at the consequence of the different approaches -----##\n",
    "#100 prospect of 10mm each, correlation of 30%, likelihood: 10% each- expected capital raised: 100mm\n",
    "num_prospect = 100\n",
    "pool_size = 1000            #1bn\n",
    "rho = .3\n",
    "prob = .1\n",
    "successprob = np.full((num_prospect, 1), prob)\n",
    "issuerweights = np.full(num_prospect, 1/num_prospect)\n",
    "zero_recov = np.full(num_prospect, 0.0)\n",
    "bins = [0,0.1, 50, 100, 150, 200, 250, 300, 350, 400, 450, 500, pool_size]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = mk.run_scenario(pool_size, rho, successprob, issuerweights, zero_recov)\n",
    "ax = mk.plot_prob_over(df)\n",
    "mk.plot_scenarios(df, bins)\n",
    "mk.stats(df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#40 prospect of 50mm each, correlation of 80%, likelihood: 5% each - expected capital raised: 100mm\n",
    "num_prospect = 40\n",
    "pool_size = 2000\n",
    "rho = .6 \n",
    "prob = .05\n",
    "successprob = np.full((num_prospect, 1), prob)\n",
    "issuerweights = np.full(num_prospect, 1/num_prospect)\n",
    "zero_recov = np.full(num_prospect, 0.0)\n",
    "bins = [0,0.1, 50, 100, 150, 200, 250, 300, 350, 400, 450, 500, pool_size]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = mk.run_scenario(pool_size, rho, successprob, issuerweights, zero_recov)\n",
    "mk.plot_prob_over(df)\n",
    "mk.plot_scenarios(df, bins)\n",
    "mk.stats(df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Current pipeline\n",
    "#\tLikelihood\t\t\t\n",
    "#size\t1\t2\t3\t4\n",
    "#25\t0\t1\t1\t6\n",
    "#10\t0\t0\t1\t12\n",
    "#5\t4\t0\t2\t19\n",
    "#2\t2\t0\t0\t16\n",
    "#35% correlation\n",
    "num = (4, 2, 1, 1, 1, 2, 6, 12, 19, 16)\n",
    "size = (5, 2, 25, 25, 10, 5, 25, 10, 5, 2)\n",
    "prob = (.5, .5, .35, .25, .25, .25, .05, .05, .05, .05)\n",
    "rho = .25\n",
    "zero_recov = np.full(sum(num), 0.0)\n",
    "pool_size = sum(n*s for n,s in zip(num, size))\n",
    "successprob, issuerweights = [], []\n",
    "for n, s, p in zip(num,size,prob):\n",
    "    successprob.append(np.full((n, 1), p)) \n",
    "    issuerweights.append(np.full(n, s/pool_size))\n",
    "successprob = np.concatenate(successprob)\n",
    "issuerweights = np.concatenate(issuerweights)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "bins = [0,0.1, 50, 100, 150, 200, 250, 300, 350, 400, 450, pool_size]\n",
    "pool_size, sum(num)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = mk.run_scenario(pool_size, rho, successprob, issuerweights, zero_recov)\n",
    "mk.plot_prob_over(df)\n",
    "mk.plot_scenarios(df, bins)\n",
    "mk.stats(df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Current LMCG pipeline\n",
    "#LMCG\tLikelihood\t\t\t\n",
    "#size\t1\t2\t3\t4\n",
    "#25\t0\t0\t0\t2\n",
    "#10\t0\t0\t0\t2\n",
    "#5\t0\t0\t1\t2\n",
    "#2\t1\t0\t0\t1\n",
    "#50% correlation\n",
    "num = (1,1,3,2,2,1)\n",
    "size = (2,5,25,10,5,2)\n",
    "prob = (.5, .25, .05, .05, .05, .05)\n",
    "rho = .6\n",
    "zero_recov = np.full(sum(num), 0.0)\n",
    "pool_size = sum(n*s for n,s in zip(num, size))\n",
    "successprob, issuerweights = [], []\n",
    "for n, s, p in zip(num,size,prob):\n",
    "    successprob.append(np.full((n, 1), p)) \n",
    "    issuerweights.append(np.full(n, s/pool_size))\n",
    "successprob = np.concatenate(successprob)\n",
    "issuerweights = np.concatenate(issuerweights)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "bins = [0,0.1, 10, 20, 30, 40, 50, 60, 70, 80, 90, 100, 110, pool_size]\n",
    "pool_size, sum(num)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = mk.run_scenario(pool_size, rho, successprob, issuerweights, zero_recov)\n",
    "mk.plot_prob_over(df)\n",
    "mk.plot_scenarios(df, bins)\n",
    "mk.stats(df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Current Agecroft pipeline\n",
    "#Agecroft\tLikelihood\t\t\t\n",
    "#size\t1\t2\t3\t4\n",
    "#25\t0\t1\t1\t3\n",
    "#10\t0\t0\t1\t10\n",
    "#5\t4\t0\t1\t17\n",
    "#2\t1\t0\t0\t15\n",
    "#30% correlation\n",
    "num = (4,1,1,1,1,1,3,10,17,15)\n",
    "size = (5,2,25,25,10,5,25,10,5,2)\n",
    "prob = (.5, .5,.35,.25,.25,.25,.05,.05,.05,.05)\n",
    "rho = .2\n",
    "zero_recov = np.full(sum(num), 0.0)\n",
    "pool_size = sum(n*s for n,s in zip(num, size))\n",
    "successprob, issuerweights = [], []\n",
    "for n, s, p in zip(num,size,prob):\n",
    "    successprob.append(np.full((n, 1), p)) \n",
    "    issuerweights.append(np.full(n, s/pool_size))\n",
    "successprob = np.concatenate(successprob)\n",
    "issuerweights = np.concatenate(issuerweights)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "bins = [0,0.1, 50, 100, 150, 200, 250, 300, 350, pool_size]\n",
    "pool_size, sum(num)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = mk.run_scenario(pool_size, rho, successprob, issuerweights, zero_recov)\n",
    "mk.plot_prob_over(df)\n",
    "mk.plot_scenarios(df, bins)\n",
    "mk.stats(df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "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": "3.6.1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}