{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "# Bitcoin Sentiment-Price Correlation\n", "## Inputs\n", "We use the average polarity scores and the closing prices of Bitcoin from August 8 to September 7, 2018." ] }, { "cell_type": "code", "execution_count": 1, "metadata": {}, "outputs": [], "source": [ "sentiments = [0.1148378151260504,0.10054880187025132,0.1320964596633778,0.13114851605087824,0.12055378205128203,0.0737112239142696,0.08774400221116639,0.10814579489962019,0.12714464882943144,0.12790344009489915,0.12617892791127544,0.11751586287042418,0.11101584158415842,0.07804477611940298,0.0945551724137931,0.09388588503130335,0.11590802248339295,0.1265067281606077,0.11124482225656877,0.11794166666666667,0.12069146775012697,0.16124164256795834,0.11008077544426494,0.12266835016835016,0.1107218982275586,0.1354550684931507,0.11982024866785078,0.13218563411896744,0.09573690415171705,0.09506849315068491,0.12344048467569496]\n", "prices = [6543.24,6153.41,6091.14,6091.14,6263.2,6199.6,6274.22,6323.81,6591.16,6405.71,6502.18,6269.9,6491.11,6366.13,6538.95,6708.96,6749.56,6720.6,6915.73,7091.38,7052,6998.76,7026.96,7203.46,7301.26,7270.05,7369.86,6705.03,6515.42,6426.33,6427.89]" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Graphical Analysis" ] }, { "cell_type": "code", "execution_count": 2, "metadata": {}, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "import matplotlib.pyplot as plt\n", "%matplotlib inline\n", "\n", "plt.figure(1)\n", "plt.suptitle('Sentiments and Prices')\n", "\n", "# Sentiments boxplot\n", "plt.subplot(121)\n", "plt.boxplot(sentiments)\n", "plt.ylabel('Sentiment (polarity)')\n", "plt.xticks([])\n", "\n", "# Prices boxplot\n", "plt.subplot(122)\n", "plt.boxplot(prices)\n", "plt.ylabel('Price (USD)')\n", "plt.xticks([])\n", "\n", "plt.tight_layout(rect=[0, 0.03, 1, 0.95])\n", "plt.show()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "We see one outlier (*approximately 0.16*) in our sentiment dataset." ] }, { "cell_type": "code", "execution_count": 3, "metadata": {}, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "plt.figure(2)\n", "plt.scatter(x=sentiments, y=prices)\n", "plt.suptitle('$BTC Sentiment v. Price\\n# obs = 31')\n", "plt.ylabel('Price (USD)')\n", "plt.xlabel('Sentiment (polarity)')\n", "plt.show()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Regression\n" ] }, { "cell_type": "code", "execution_count": 4, "metadata": {}, "outputs": [ { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAAY4AAAEjCAYAAAAlhuZMAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMS4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvNQv5yAAAIABJREFUeJzt3Xvc3OOd//HXWxIklDsqLQlpqIilIbhRUlqU0J9FsUVtBbvVbvWgq0pqt1Rb567Sqq6tONU5JY1DpVFlu20dEnEMqYhDDg6JCEUqp8/vj+81vee+M3PPPfc9x/t+Px+PeWTm+h7m8x23+cz1vU6KCMzMzLpqrXoHYGZmzcWJw8zMyuLEYWZmZXHiMDOzsjhxmJlZWZw4zMysLE4cVhGSzpI0ot5x1Iqk4ZLekdSv3rHUk6TvSPpFveOw2nLisKqR9ClJq9MX7DuSFkj6Xto2PK/8HUkh6d2813um/XaVdLekpZKWSHpY0vFF3m9tST+SND+d4wVJF1foWl6U9Onc64h4OSLWj4hVlTh/mbGEpK2qdO6rJS1Pn98SSdMkbVNs/4g4JyL+tRqxWONy4rAekfRxSQ8D3wYelfR/ktbP22Vh+oJdH/gE8C+SDs374s1tA9ghr+wPknYH7gMeALYCPgj8G3BgkXAmAK3ArsAHgL2BmZW+5j7ggvTfZDPgdeDqQjtJ6l/LoKxxOHFYT10D3AJcBOwOnAUU/BUeES8AfwK27eK5LwSuiYjzI2JxZGZExOeK7L8LcHtELEz7vhgR1+Y2Shoq6VeSFqXayNfztp0l6RZJ10r6q6SnJbWmbdcBw4E70i/xb0sakX7590/73C/pB5L+lPa5Q9IHJV0v6W1Jj+TfypO0Tfo1v0TSbEmfy9t2taTLJN2VYnlI0kfTtv9Nuz2e3ufI/A9A0jqpdvaxvLIhkpZJ+lAXP3cAIuI94AbgY3mf0SRJv5T0NnBcKvtl3nt9In0GSyXNk3RcXlwXSXpZ0muSfi5pYDnxWONw4rCe+jDwZ2A18H5E3BsRywrtKGkkMBZ4sNRJJQ0iS0STyojlQeDfJX1F0mhJyjvfWsAdwOPAMGBf4GRJ4/KOPxi4CWgBpgA/BYiILwAvA/+YakMXFHn/o4AvpPN/lOxzuQrYCHgGODPFsh4wjexL+UPA0cDPJG2Xd66jge8Bg4E5wA9TLHul7bna2c35AUTE+8Bt6ficzwEPRMTrReIuKNUcj6F9re0Qsv8mLcD1HfYfDvwG+AkwBBgDPJY2nw9sncq2IvuMvltOPNY4nDispy4AJpN9Of2TpA932D40/fp8G/gL8BDwf10472Cyv89XyojlXLIvqGOA6cACSePTtl2AIRFxdkQsj4i5wP+Qfdnn/F9E3J3aLa4DdijjvQGuiojnI+Itsi/Q51MiXQncCuyY9jsIeDEiroqIlRHxKPAr4Ii8c90WEQ+nY68n+8Ltqhtonzg+n8q66luSlpIlrPWB4/K2/TkiJkfE6gI/EI4B7o2IGyNiRUS8ERGPpQT+ReCbEbEkIv4KnEP7z96aiBOH9UhEnAPsQfaLehwwW9IuebssjIiWiNiA7FfqMrLbW6W8SVaL2bSMWFZFxGURMTa91w+BiZL+AfgIbUlsafpi/A5ZjSnn1bzn7wHrlnkf/7W858sKvM615XwE2K1DLMcAm3QSS367USn3AQMl7SbpI2RJ5/Yyjr8o/TfbJCIOjojn87bN6+S4zYHnC5QPAQYBM/Ku955Ubk3IjVvWYxHxnKQnyRpRJwDjgUcK7PeWpBuAmztuK7Dve5L+DBwO/L4bMS0DLlPWi2tbsi+8FyJiZLnnyp2ym8cVMo/s1tF+FTzn30XEakm3kNU6XgPuTL/yK3L6TrbNI+uY0NFissS5XUQsqFAcVkeucViPSPqipHXSy3XJ7l+/VmTf9cluTzzdxdN/m6wB9lRJH0zn2EHSTUXOf7KyLsADJfVPt6k+QHaP/mHgbUmnpe39JH2sQ+2oM68BW3Zx31LuBLaW9AVJA9Jjl1QzqlQsNwBHktVkyrlN1RPXA5+W9Ln0+X9Q0piIWE12W/DiXAO9pGEd2pesiThxWE/tAcwGvgH8gewW04/ztg9NvX/eAV4iayg+pisnjog/Afukx1xJS4ArgLuLHLIM+BHZbZ7FwEnA4RExN7Vb/CPZbZsX0vZfABt28TrPBf4j3Wr5VhePKSj9+t+fLIkuTPGeD6zT2XF5zgKuSbEU7GEWEQ8B7wJDydpbgHbjZ4Z3/woKi4iXgc8ApwBLyBrGc+1Ep5G1mTyY2rvuBUZVOgarDXkhJ6sESWcBV0fEi3UOxcyqzDUOMzMri2scZmZWFtc4zMysLE4cZmZWFicOsy5K81F5Jljr85w4rNdTNhX7SElbSnq03vF0l6TfK5ug8W1Jj0s6JG/bppKmSFqobPLFEfWL1Ho7Jw7r1SQNIJviYw6wM9C0iYNsrMymafqWE4FfSspNybKabBqPw+sVnPUdThzW230MmBVZ98FWSiQOSXsomwL9rfTvHh12+Wiqwbwl6deSNkrHrZumG38jDcx7pMCEjz0SEU+kSQ8hm/pjANn8UETEaxHxMwpM9WJWaU4c1itJOj5NpvdHYPf0/BTg/PTFvkWBYzYC7gIuJVs06r+Au3LTnSTHAieQjchemfaFbH6uDcm+yD8IfJlsJHuh2O7Mn+Cww+POEtd1p6S/kc0yfD/ZLMBmNeXEYb1SmrK8BZgBfBzYHngK2CDN/PpCgcP+H/BcRFyXpju/EXiWbKqSnOsi4qmIeBf4T+BzytYdX0GWMLZKs/TOiIi3i8R2UIqh0OOgEtd1ENn8W58BpqZ5oMxqyonDeh1JG6Vf72+RzaV1P9l8WqOANyWdXOTQoWTzaeV7iWzRoZx5HbYNADYmW79jKnBTaqC+ILWvVFxa6+I3wDhJB1fjPcw648RhvU5aLKgF+BLwi/T8HrIV/Foi4sdFDl1I1pCebziQPxX45h22rQAWpy/z70XEtmTJ6iCy21prkPSb3MSPBR6/KXRMEf3JVho0qyknDuvN8ntR7Uh226ozd5NNd/75NC34kWRreeS3O/yzpG2VLW17NjApIlZJ2lvZcrX9gLfJEkqxtdcPTMu+FnocWOgYZWuUH5imhB8g6Z+BvYAH8vZZl7YZdtdJr80qzonDerOdgUdT4/aqiHizs50j4g2ymsIpwBtk64EcFBGL83a7jmzBqlfJ1h/5eirfhGwt7rfJVkN8APhlxa4ERDad+uvAIrKuuUemZWdzlgHvpOfPUqRx3qynPMmhmZmVxTUOMzMrixOHmZmVpX+9A6iGjTfeOEaMGFHvMMzMmsqMGTMWR8SQUvv1ysQxYsQIpk/3gFozs3JI6jiOqaCq3aqSNErSY3mPt/MHXkn6VprFc+P0WpIulTRH0hOSdsrbd7yk59JjfLViNjOz0qpW44iI2cAYgNS3fQFwe3q9ObAf8HLeIQcCI9NjN+ByYLc0f9CZZBPUBTBD0pRSXSvNzKw6atU4vi/wfETkqkEXk/WRz+8LfAhwbWQeBFrSlNHjgGlpNPCbwDTggBrFbWZmHdQqcRwF3AiQ5tZZEBGPd9hnGO3nAZqfyoqVm5lZHVS9cVzS2sDBwIQ0TcMZwP6Fdi1QFp2Ud3yfE8kWt2H48OHdjtfMzDpXixrHgcCjEfEa2YRsWwCPS3oR2IxsSohNyGoS+RPIbUY26Vyx8nYi4oqIaI2I1iFDSvYmM2sok2cuYOx597HF6Xcx9rz7mDxzQemDzOqkFonjaNJtqoh4MiI+FBEjImIEWVLYKSJeBaYAx6beVR8H3oqIV8imqt5f0mBJg8lqK1NrELdZTUyeuYAJtz3JgqXLCGDB0mVMuO1JJw9rWFVNHOnW1H7AbV3Y/W5gLtna0P8DfAWyKbKB75MtifkIcHYqM+sVLpw6m2Ur2k+ku2zFKi6cOrtOEZl1rqptHBHxHtmqaMW2j8h7HsBJRfabCEysdHxmjWDh0sKT2BYrN6s3z1VlVmdDWwaWVW5Wb04cZnV26rhRDBzQr13ZwAH9OHXcqDpFZNa5XjlXlVkzOXTHbFjShVNns3DpMoa2DOTUcaP+Xm7WaJw4zBrAoTsOc6KwpuFbVWZmVhYnDjMzK4sTh5mZlcVtHGZFTJ65wA3WZgU4cZgVkJsGJDeiOzcNCODkYX2eb1WZFeBpQMyKc+IwK8DTgJgV58RhVoCnATErzonDrABPA1Ka1xDpu9w4blaApwHpnDsP9G1OHGZFeBqQ4jrrPODPrPfzrSozK5s7D/RtThxmVjZ3HujbnDjMrGzuPNC3uY3DzMrmzgN9mxOHmXWLOw/0Xb5VZWZmZXHiMDOzslQtcUgaJemxvMfbkk6WdKGkZyU9Iel2SS15x0yQNEfSbEnj8soPSGVzJJ1erZjNGpVHaVsjqVriiIjZETEmIsYAOwPvAbcD04CPRcT2wF+ACQCStgWOArYDDgB+JqmfpH7AZcCBwLbA0Wlfsz4hN0p7wdJlBG2jtJ08rF5qdatqX+D5iHgpIn4bEStT+YPAZun5IcBNEfF+RLwAzAF2TY85ETE3IpYDN6V9zfoET/FujaZWieMo4MYC5ScAv0nPhwHz8rbNT2XFys36BI/StkZT9cQhaW3gYODWDuVnACuB63NFBQ6PTso7vs+JkqZLmr5o0aKeBW3WQDxK2xpNLWocBwKPRsRruQJJ44GDgGMiIpcE5gOb5x23GbCwk/J2IuKKiGiNiNYhQ4ZU+BLM6sejtK3R1CJxHE3ebSpJBwCnAQdHxHt5+00BjpK0jqQtgJHAw8AjwEhJW6Tay1FpX7M+4dAdh3HuYaMZ1jIQAcNaBnLuYaM9+M7qpqojxyUNAvYDvpRX/FNgHWCaJIAHI+LLEfG0pFuAWWS3sE6KiFXpPF8FpgL9gIkR8XQ14zZrNB6lbY1EbXeKeo/W1taYPn16vcMwM2sqkmZERGup/Txy3MzMyuLEYWZmZXHiMDOzsjhxmJlZWZw4zMysLE4cZmZWFicOMzMrixOHmZmVxWuOm9XQ5JkLuHDqbBYuXcbQloGcOm6UR4Rb03HiMKuR3IJMubU1cgsyAU4e1lR8q8qsRrwgk/UWThxmNeIFmay3cOIwqxEvyGS9hROHWY14QSbrLdw4blYjuQbwznpVudeVNQMnDrMa6mxBJve6smbhW1VmDcK9rqxZOHGYNQj3urJm4VtVZgXUo61haMtAFhRIEu51ZY3GNQ6zDnJtDQuWLiNoa2uYPHNBVd/Xva6sWThxmHVQr7aGQ3ccxrmHjWZYy0AEDGsZyLmHjXbDuDUc36oy66CebQ2d9boyaxRVq3FIGiXpsbzH25JOlrSRpGmSnkv/Dk77S9KlkuZIekLSTnnnGp/2f07S+GrFbAYe4W1WStUSR0TMjogxETEG2Bl4D7gdOB34XUSMBH6XXgMcCIxMjxOBywEkbQScCewG7AqcmUs2ZtXgtgarpMkzFzD2vPvY4vS7GHvefVVvK6uFWrVx7As8HxEvAYcA16Tya4BD0/NDgGsj8yDQImlTYBwwLSKWRMSbwDTggBrFbX2Q2xqsUurV0aLaatXGcRRwY3r+4Yh4BSAiXpH0oVQ+DJiXd8z8VFasvB1JJ5LVVBg+fHhFg7e+x20NVgmddbRo5r+vqtc4JK0NHAzcWmrXAmXRSXn7gogrIqI1IlqHDBlSfqBmZhXWWwd11uJW1YHAoxHxWnr9WroFRfr39VQ+H9g877jNgIWdlJuZNbTe2tGiFonjaNpuUwFMAXI9o8YDv84rPzb1rvo48Fa6pTUV2F/S4NQovn8qMzNraL21o0VV2zgkDQL2A76UV3wecIukfwFeBv4pld8NfAaYQ9YD63iAiFgi6fvAI2m/syNiSTXjNjOrhK5Mpd+MFLFGc0HTa21tjenTp9c7DDOzpiJpRkS0ltrPU46YmVlZnDjMzKwsThxmZlYWJw4zs2qbOxcWLap3FBXj2XHNSqjHok7WCyxcCCecAFPT6IEPfADefru+MVWIE4dZJ3JzDeWmjcjNNQQ4edia3nwTTjoJbrxxzW2TJ9c+nirxrSqzTtRrUSdrIu+9B1//Okiw0Ubtk8aFF8LKlRAB++xTvxgrzInDrBO9da4h66EVK+Css7Jksd568JOftG07/XT429+yZPGtb0G/fkVP06x8q8qsE0NbBrKgQJJo9rmGrBtWr4bLLstqFx198Yvwox9l7Rh9QFk1DknrSep96dOsiN4615B1UQTccENWs+jXr33SOOwweP31bJ8rrugzSQNK1DgkrUW2lsYxwC7A+8A6khaRzS11RUQ8V/Uozeqkt841ZCXccw8ceii8/3778k9+Eq65Bj7ykfrE1SBK3ar6PXAvMAF4KiJWw9+Xc90bOE/S7RHxy+qGaVY/XtSpj3jwQTj88Kwbbb7Ro7MG7+22q09cDahU4vh0RKzoWJhmp/0V8CtJA6oSmZlZtc2aBUcdBU8+2b586FCYNAl2370+cTW4ThNHLmlIGg1sk4qfiYinOu5jZtYUXn4Zjj0WHnigffk668Dtt8OBB9YnriZSqo1jQ7KFljYHniBbxnW0pJeBQyKidwyDNLPebfFi+NKX4Lbb1tx2/fVw9NFZA7h1SaleVd8HpgMjI+KzEXEoMJJsUaUfVjs4M7Nue+cdOPHELCEMGdI+aVx6KaxalfWI+vznnTTKVLKNA9g+1ygOEBGrJX0HeLL4YWZmdbB8OZx5Jpx33prbzjwTzjgDBrhZtqdKJY7lEbGyY2FErJT0fqEDzMxqatUquPhiOPXUNbd99atw/vkwaFDt4+rFSiWOdSXtSNa2kU/AOtUJycyshAi46ir4l39Zc9vRR8NPf5rNG2VVUSpxvAr8VyfbzMxq59e/zgbmdbT//jBxIgzzeJtaKNUd91M1isPMrLD//V/47GdhyZL25TvvnE0HsvXW9YmrD+u0V5WkXSRtkvf6WEm/lnRpGj1uZlZ5jz+eJQQpm+YjlzS22AIeeSS7VTV9upNGnZTqjvvfwHIASXsB5wHXAm8BV5Q6uaQWSZMkPSvpGUm7Sxoj6UFJj0maLmnXtK9SQpoj6QlJO+WdZ7yk59JjfHcv1swa2PPPZyO1JRgzBp5L0+BtsAH87ndZspg7F1pb6xunlWzj6JemFwE4kmxSw9xUI4914fyXAPdExBGS1gYGAbcA34uI30j6DHAB8CngQLIxIiOB3YDLgd1SzeZMoBUIYIakKRHxZjkXamYN6NVX4V//Fe66a81tt94KRxxR+5ispFI1jn6ScsllX+C+vG2lRp1vAOwFXAkQEcsjYinZl/8GabcNgdyMYocA10bmQaBF0qbAOGBaRCxJyWIacECXrs7MGs9bb8H48VnNYtNN2yeN//7vbN2LCCeNBlaqxnEj8ICkxcAy4A8AkrYiu13VmS2BRcBVknYAZgDfAE4Gpkq6iCxx7ZH2HwbMyzt+fiorVt6OpBOBEwGGDx9eIjQzq6lly+A734Ef/3jNbeecA9/+dq9cKa+36rTGERE/BE4BrgY+ERGRd9zXSpy7P7ATcHlE7Ai8C5wO/BvwzYjYHPgmqUbCmmNFIKudFCvvGOsVEdEaEa1DhgwpEZqZVd3KlfCDH2Q1i0GD2ieNU07JkkkETJjgpNFkSvWq2gj4C/AA2QJOgyUpIv4SEY+WOPd8YH5EPJReTyJLJOOB3KQxtwK75u2/ed7xm5HdxipWbmaNJgIuvzxLFgMGwH/+Z9u2446DpUuzfS66CNZdt25hWs+UauOYQTbJ4fT0/FHgdUn3ShrR2YER8SowT1Jujc19gVlkX/qfTGX7ALkVBKcAx6beVR8H3oqIV4CpwP4paQ0G9k9lZtYobrklSxZrrQVf+Upb+cEHZw3guZHeG25YvxitYkoNANyiULmkw4CfU7qR+mvA9alH1VzgeLJp2i9Jje5/I7VLkC1F+xlgDvBe2peIWCLp+2Qz8gKcndfTy8zq5bTT4IIL1izfYw+47jrYcsvax2Q1obZmizIPlB6NiJ1K71l7ra2tMX369HqHYX3Y5JkLeuc65Zdf3r5GkbPNNnDzzbD99rWPySpG0oyIKDlQplSvqmInX5/St7nM+qTJMxcw4bYnWbZiFQALli5jwm3ZKgRNmTymTIFDDim87ZxzssZt61NKjcX49wLFg4GDgZ9WJSKzJnfh1Nl/Txo5y1as4sKps5sncVx5ZTYwr5AvfhGuKDlxhPVipWocH+jwOshmxf3niPBCTmYFLFy6rKzyhjF1KhxQpNly772zaT+8Up5RunH8e7UKxKy3GNoykAUFksTQloF1iKaEWbNgu+2Kb1+xAvp364629WKlxnFcIeljRbatJ+kEScdUJzSz5nTquFEMHNB+QNvAAf04ddyoIkdU1uSZCxh73n1scfpdjD3vPibPXNB+h8WLs5qDVDhp5LrPRjhpWEGl/ip+BnxX0mjgKbIpRNYlm4hwA2AicH1VI7Sm1Wt7FpWQu8Z6XHuxhvm1lr/PwR//aPEDH3/cPaKsy7rUHTf1omoFNiWbs+qZiJhd5di6zd1x66/jFxhkv7rPPWx0n0ge9TL2vPvabpNF8OIF/1h85ylT4B/X3N5XE75VuDtuRLwD3N/ToKzv6BU9i5rQwqXLePH8g4rvcNFF2TxRRfS6rsRWFR6LYVXRtD2LmlVqs3ihQNJ4YpOtGHtuWgipk6QBnSd8sxy3fFlVNFXPomZVomvsiNPuBNItwi42zDvhW1eUVeOQtF61ArHepd49i3qtnXdu6xFVSASTH53P2HN/h4BhLQPLalcqltid8C1fl2ockvYAfgGsDwxPCzN9KSIKTFpjVt+eRb3OccfBNdcU396hg8uhOw7r9ud86rhRBTs1OOFbvq7eqrqYbAnXKQAR8bikvaoWlfUKPfkC6/OuugpOOKH49r/9DdZZp+Jv64RvXdHlNo6ImKf21eNVxfY1s26YMQNaO+kJ+dJLUINlkZ3wrZSuJo556XZVpLU1vg48U72wzPqIN9+EjTYqvv3ee2HffWsXj1kXdLVx/MvAScAwsqVcx6TXZlauVavaGrgLJY1zz22b8sNJwxpQVwcALgY8J5VZT3TWffbYYztvADdrIF2qcUi6RlJL3uvBkiZWLyyzXiJXsyiUND70obaahZOGNZGutnFsHxFLcy8i4k1JO1YpJrPmVmrNim4u12zWKLraxrGWpMG5F5I2wqPOzdr0719yYN7fH2ZNrquJ40fAnyR9X9L3gT8BF1QvLLMmcNhhbcliVYHe6U4W1kt1KXFExLXA4cBrwOvAYRFxXTUDM2tIl1zSlixuv33N7StXOllYr1dqBcAN0r8bka01fgPZwk2vprJOSWqRNEnSs5KekbR7Kv+apNmSnpZ0Qd7+EyTNSdvG5ZUfkMrmSDq9e5dq1k2PPNKWLE4+ec3tb7zRliz69Vtzu1kvU6qd4gbgIGAGkP8TSun1liWOvwS4JyKOSAMHB0naGziErMH9fUkfApC0LXAUsB0wFLhX0tbpPJcB+5GNIXlE0pSImNXVizQr29KlMHhw8e2PPQY77FC7eMwaSKeJIyIOUjbPyCcj4uVyTpxqK3sBx6VzLQeWS/o34LyIeD+Vv54OOQS4KZW/IGkOsGvaNici5qbz3pT2deKwylq5EgYMKL79xhvhqKNqF08PeSU/q5aSbRyRrS1b4GZuSVuSrVF+laSZkn6RpmXfGthT0kOSHpC0S9p/GDAv7/j5qaxYeTuSTpQ0XdL0RYsWdSNc67Nyt6EKJY3LLmu7DdVkSWPCbU+yYOkygraV/CbPXFDv0KwX6GqvqgfzvuC7qj+wE3B5ROwIvAucnsoHAx8HTgVuSbWaQv0Yo5Py9gURV0REa0S0DhkypMxQrc/pbGDe977Xliy+0pwrB3glP6umro7F2Bv4sqQXyRKAyCoj23dyzHxgfkQ8lF5PIksc84HbUk3mYUmrgY1T+eZ5x28GLEzPi5Wbdd3WW8NzzxXedsghMHlybeOpIq/kZ9XU1cRxYLknjohXJc2TNCoiZgP7krVLPA/sA9yfGr/XBhaTrfVxg6T/ImscHwk8TJakRkraAlhA1oD++XLjsT7q2GPhuiI9xz/8YXj11drGUyNeuteqqdPEIWldsplxtwKeBK6MiJVlnP9rwPWpR9Vc4HiyGstESU8By4HxqfbxtKRbyJLLSuCkiFiV4vgqMBXoB0yMiKfLiMH6mosuglNPLb69D4yx8Ep+Vk2KTv4nknQzsAL4A1mt46WI+EaNYuu21tbWmD59er3DsFq64w44+ODi21evLj2HVC/jXlVWLkkzIqKT1cQypW5VbRsRo9MJryS7dWTWGJ54ovOxFO+/D2uvXbNwGu2L2iv5WbWUShwrck8iYqX62C82q4yKfqG+9hpssknx7a+/DnXoVZfr/pq7NZTr/gr4y9t6nVLdcXeQ9HZ6/BXYPvdc0tu1CNCaW0XGEyxf3tZ1tlDSePrptu6zdeqK7e6v1pd0mjgiol9EbJAeH4iI/nnPN6hVkNa8uv2FGtGWLNZZZ83td9/dliy23baCEXePu79aX9LVAYBm3VLsi3PB0mWMPe++NWseQ4dmyWKtAn+akya1JYsDy+4hXlXFurm6+6v1Rk4cVlWdfXHmblu9uuen22oXr7zSfqcLLmhLFocfXuVou+/UcaMYOKD9zLju/mq9lVfxs6oqNJ4A4LT7r+bfHppU+KDjj4eJzbWkfa4BvN69qhqtZ5f1Tk4cVlX5X6ijpj/AxF+dXXjHnXaCGTNqGFnl1bv7q3t2Wa04cVh1/e53HPrpT3Nokc0jTruTYS0D+ePp+9Q0rK5qpl/wnXVEaNSYrTk5cVjlzZ8Pm29edPOWp/6a1Wtl7QGN3A7QbL/g3bPLasWN41YZb73V1sBdKGm88w5EMPnR+Wy60foIGNYykHMPG92QX8LQfGMz3LPLasU1Duu+5cvhgx/MkkIhb74JLS3tiurdDlCOZvsF74kNrVZc47DyrF6dNWTnBuZ1TBovvdTWfbZD0mg2zfYL/tAdh3HuYaMZ1jKwKWp01rxc47CuOfpouOmmwtsee6zzyQabVDP+gm+mGp01L9c4rLgzzmhrt+iYNH7727b2RpnbAAAQBElEQVSaRS9MGuBf8GbFuMZh7V1xBXzpS4W3XXNNtqJeH+Jf8GZrco3D4M4722oWHZPG2We31Sz6WNIws8Jc4+irHnkEdt218LYTToArr6xtPGbWNJw4+pK5c+GjHy28bc894YEHqra8ajONwG5k/hytEThx9HaLFxdf3OjDH4Z582DAgKqG0GwjsBuVP0drFG7j6I2WLWtrsyiUNN5+O2uzePXVqicNaL4R2I3Kn6M1Ctc4eotVq2DUKHj++cLbX3ml87W6q6jZRmA3qu5+jr69ZZVW1RqHpBZJkyQ9K+kZSbvnbfuWpJC0cXotSZdKmiPpCUk75e07XtJz6TG+mjE3ldxKeBL0779m0pg1q61HVJ2SBjTfCOxG1Z3PsSJrvpt1UO1bVZcA90TENsAOwDMAkjYH9gNeztv3QGBkepwIXJ723Qg4E9gN2BU4U9LgKsfd2L7xjbblVe+5p/22P/yhLVn8wz/UJ74OvDpeZXTnc/TtLauGqiUOSRsAewFXAkTE8ohYmjZfDHwbiLxDDgGujcyDQIukTYFxwLSIWBIRbwLTgAOqFXfDuvXWtnaLSy9tv+2WW9qSxSc+UZ/4OuER2JXRnc/RtwmtGqrZxrElsAi4StIOwAzgG8C+wIKIeFztu34OA+blvZ6fyoqVtyPpRLKaCsOHD6/cVdTTrFlw5JHw1FNrbrv4Yjj55NrH1E0egV0Z5X6OQ1sGsqBAkvBtQuuJat6q6g/sBFweETsC7wJnAWcA3y2wf6EBBNFJefuCiCsiojUiWocU637aDF56CT71qaxmsd127ZPGD37QVrNooqRh9ePbhFYN1Uwc84H5EfFQej2JLJFsATwu6UVgM+BRSZuk/fNXANoMWNhJee+xeDEcfniWLEaMyAbi5Vx/PZNnzGPsub9ji7+OYex597lh07rMtwmtGqp2qyoiXpU0T9KoiJhNdovq0YjYN7dPSh6tEbFY0hTgq5JuImsIfysiXpE0FTgnr0F8f2BCteKumb/+FU45Bf7nf9bcdumlcNJJsNZaHvRlPebbhFZp1R7H8TXgeklrA3OB4zvZ927gM8Ac4L3cvhGxRNL3gUfSfmdHxJLqhVwDb7wBG2/cvuzMM7NpzDsMyOusV4y/DMysHqqaOCLiMaC1k+0j8p4HcFKR/SYCEysdX91suCF84QvZv+efD4MGFd3VvWLMrNF45Hg99O8P117bpV3dK8bMGo3nqmpw7hVjZo3GNY4Gl2vH8FxDZtYonDiagHvFVI8nADQrnxOH9Vnu6mzWPW7jsD7LEwCadY8Th/VZ7ups1j1OHNZneZ0Qs+5x4rA+y12dzbrHjePWZ7mrs1n3OHFYn+auzmblc+KwNXhsQ3X4c7XewonD2vHYhuro6ufq5GLNwI3j1o7HNlRHVz7XXHJZsHQZQVty8cJd1micOKwdj22ojq58rk7a1iycOKwdj22ojq58rk7a1iycOKwdj22ojq58rk7a1iycOKydQ3ccxrmHjWZYy0AEDGsZyLmHjXYDbQ915XN10rZmoWzF1t6ltbU1pk+fXu8wzMrmXlVWT5JmRETR5b5z3B3XrIF4QKI1A9+qMjOzsjhxmJlZWaqaOCS1SJok6VlJz0jaXdKF6fUTkm6X1JK3/wRJcyTNljQur/yAVDZH0unVjLkWJs9cwNjz7mOL0+9i7Hn3eYCXmTWVatc4LgHuiYhtgB2AZ4BpwMciYnvgL8AEAEnbAkcB2wEHAD+T1E9SP+Ay4EBgW+DotG9T8uhgM2t2VUsckjYA9gKuBIiI5RGxNCJ+GxEr024PApul54cAN0XE+xHxAjAH2DU95kTE3IhYDtyU9m1KHh1sZs2umjWOLYFFwFWSZkr6haT1OuxzAvCb9HwYMC9v2/xUVqy8HUknSpouafqiRYsqdQ0V59HBZtbsqpk4+gM7AZdHxI7Au8Df2ycknQGsBK7PFRU4R3RS3r4g4oqIaI2I1iFDhvQ09qrx6GAza3bVTBzzgfkR8VB6PYkskSBpPHAQcEy0jUCcD2yed/xmwMJOypuSRwebWbOrWuKIiFeBeZJy34j7ArMkHQCcBhwcEe/lHTIFOErSOpK2AEYCDwOPACMlbSFpbbIG9CnVirvaPKWHmTW7ao8c/xpwffrCnwscT5YI1gGmSQJ4MCK+HBFPS7oFmEV2C+ukiFgFIOmrwFSgHzAxIp6uctxV5dHB5fNUHGaNw3NVWcPruHoeZLf3XFMzq6yuzlXlkePW8NyF2ayxOHFYw3MXZrPG4sRhDc9dmM0aixOHNTx3YTZrLF6PwxpergHcvarMGoMThzUFd2E2axy+VWVmZmVx4jAzs7L4VpWZtRuZ3zJoABHw1rIVbk+ygpw4zPq4jiPz33xvxd+35RYaA5w87O98q8qsjys0Mj+fR+lbR04cZn1cV0bge5S+5XPiMOvjujIC36P0LZ8Th1kfV2hkfj6P0reO3Dhu1sd1HJnvXlVWihOHmXlkvpXFt6rMzKwsThxmZlYWJw4zMyuLE4eZmZXFicPMzMqiiKh3DBUnaRHwErAxsLjO4dRCX7jOvnCN4OvsbZrtOj8SEUNK7dQrE0eOpOkR0VrvOKqtL1xnX7hG8HX2Nr31On2ryszMyuLEYWZmZentieOKegdQI33hOvvCNYKvs7fpldfZq9s4zMys8np7jcPMzCqsaROHpAMkzZY0R9LpBbavI+nmtP0hSSPytm0v6c+Snpb0pKR1axl7V3X3GiUNkHRNurZnJE2odezl6MJ17iXpUUkrJR3RYdt4Sc+lx/jaRV2+7l6npDF5f69PSDqytpF3XU/+W6btG0haIOmntYm4e3r4Nztc0m/T/5uz8r+bmkZENN0D6Ac8D2wJrA08DmzbYZ+vAD9Pz48Cbk7P+wNPADuk1x8E+tX7mip8jZ8HbkrPBwEvAiPqfU09uM4RwPbAtcAReeUbAXPTv4PT88H1vqYqXOfWwMj0fCjwCtBS72uq5DXmbb8EuAH4ab2vp1rXCdwP7Jeerw8Mqvc1lfto1hrHrsCciJgbEcuBm4BDOuxzCHBNej4J2FeSgP2BJyLicYCIeCMiii+4XD89ucYA1pPUHxgILAferk3YZSt5nRHxYkQ8AazucOw4YFpELImIN4FpwAG1CLobun2dEfGXiHguPV8IvA6UHKRVBz35b4mknYEPA7+tRbA90O3rlLQt0D8ipqX93omI92oUd8U0a+IYBszLez0/lRXcJyJWAm+R1S62BkLS1FSV/HYN4u2OnlzjJOBdsl+mLwMXRcSSagfcTV25zmocW2sViVXSrmS/cp+vUFyV1O1rlLQW8CPg1CrEVWk9+W+5NbBU0m2SZkq6UFLx5RcbVLMmDhUo69g9rNg+/YFPAMekfz8rad/KhlcRPbnGXYFVZLc1tgBOkbRlZcOrmK5cZzWOrbUexyppU+A64PiIWOMXewPoyTV+Bbg7IuaV3LP+enKd/YE9gW8Bu5Dd7jquMmHVTrMmjvnA5nmvNwMWFtsn3bLZEFiSyh+IiMWping3sFPVIy5fT67x88A9EbEiIl4H/gg06rQHXbnOahxbaz2KVdIGwF3Af0TEgxWOrVJ6co27A1+V9CJwEXCspPMqG17F9PRvdma6zbUSmExjfv90qlkTxyPASElbSFqbrGF4Sod9pgC5XjZHAPdF1ho1Fdhe0qD0ZftJYFaN4i5HT67xZWAfZdYDPg48W6O4y9WV6yxmKrC/pMGSBpO1X02tUpw91e3rTPvfDlwbEbdWMcae6vY1RsQxETE8IkaQ/Rq/NiLW6K3UIHryN/sIMFhSro1qHxrz+6dz9W6d7+4D+AzwF7J7vWeksrOBg9PzdYFbgTnAw8CWecf+M/A08BRwQb2vpdLXSNZT49Z0jbOAU+t9LT28zl3Ifqm9C7wBPJ137Anp+ueQ3cKp+/VU+jrT3+sK4LG8x5h6X0+l/1vmneM4GrhXVQX+Zvcj69n5JHA1sHa9r6fch0eOm5lZWZr1VpWZmdWJE4eZmZXFicPMzMrixGFmZmVx4jAzs7I4cVhTkXRG3iyxj0narZvnGSPpM3mvDy40y2klSfqUpD062X6opO/24Nx3lnnMUEmT0vN2n0cnx3xV0vHdidF6j/71DsCsqyTtDhwE7BQR70vamGzepu4YQzaa/m6AiJhC1wdxddengHeAPxXZ/m3g4CrHAGQzDUQ2YWJuyu92n0cnJpLNRHBVFcOzBucahzWTTYHFEfE+QGTTxiyEbGZVSQ9ImpEmsNw0ld8v6XxJD0v6i6Q902jfs4EjU63lSEnH5daAkHS1pMsl/V7SXEmflDQxrZ9wdS4YSfsrWyfjUUm3Slo/lb8o6Xup/ElJ26Q1F74MfDO95575FyZpa+D9iFicF8PPJf0hxX1QKl9X0lXpvDMl7d3xQ5K0q6Q/pe1/kjQqlR+X4rwD+K2kEZKeKvJ5PJcb3SxpLWXrTmwc2TQ9LyqbbNH6KCcOaya/BTZPX6Q/k/RJyBauAn5Ctu7BzmS/in+Yd1z/iNgVOBk4M7KpsL9Ltn7JmIi4ucB7DSabDuKbwB3AxcB2wOh0W2dj4D+AT0fETsB04N/zjl+cyi8HvhURLwI/By5O7/mHDu83Fni0Q9kIsilx/h/wc2ULjp0EEBGjgaOBa7TmQmTPAntFxI7pOs/J27Y7MD4i9skVFPk8fkk2ESjAp4HHc0ktXWu7xGd9i29VWdOIiHeUrdmwJ7A3cHNql5gOfAyYJgmyhXZeyTv0tvTvDLIv4664IyJC0pPAaxHxJICkp9M5NgO2Bf6Y3nNt4M9F3vOwLrzfpsCiDmW3RDYL7nOS5gLbkM3o/BOAiHhW0ktkU3Xn25AsoYwkm7V1QN62adG1KfYnAr8Gfkw2rUv+ranXUyzWRzlxWFOJbNGt+4H705f6eLIv56cjYvcih72f/l1F1//mc8esznuee90/nWtaRBxdofdcRvaFn6/jfEBB4Sm9O/o+8PuI+Gy6RXZ/3rZ3u3A8ETFP0muS9gF2o632Adkcacu6ch7rnXyrypqGpFHpV3TOGOAlYDYwJDWe59Zc367E6f4KfKAH4TwIjJW0VXrPQamdorvv+QywVYeyf0rtCx8lW7dhNvC/pC/x9H7DU3m+DYEF6flxpS+laGy/ILtldUu0XyVza7IJQq2PcuKwZrI+2S2YWZKeILtVdFa6R38EcL6kx8lmjy3a7TX5PbBtrjG43EAiYhHZl/KNKZYHKX375g6yhcPWaBwnSwg7Kt33SmYDDwC/Ab4cEX8Dfgb0S7Wtm4Hjcp0F8lwAnCvpj2S37bqi0Ocxhewz79iDaixwbxfPa72QZ8c1axCSLiFrW7k39d66MyIm1TGeVrLG/D3zynYE/j0ivlCvuKz+XOMwaxznAIPqHQRA6nTwK2BCh00bA/9Z+4iskbjGYWZmZXGNw8zMyuLEYWZmZXHiMDOzsjhxmJlZWZw4zMysLE4cZmZWlv8PneMSzWRoRscAAAAASUVORK5CYII=\n", "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "Intercept: 5894.182390072815\n", "Slope: 6453.203545109107\n", "Correlation: 0.31132136753415407\n", "R-squared: 0.09692099388333585\n", "p-value: 0.08823306157092385\n", "t-statistic: 1.7641887782716776\n" ] } ], "source": [ "from scipy import stats\n", "slope, intercept, r_value, p_value, std_err = stats.linregress(x=sentiments, y=prices)\n", "\n", "plt.figure(2)\n", "plt.scatter(sentiments, prices)\n", "plt.plot(sentiments, [intercept + slope*x for x in sentiments], 'r')\n", "plt.suptitle('$BTC Sentiment v. Price\\n# obs = 31')\n", "plt.ylabel('Price (USD)')\n", "plt.xlabel('Sentiment (polarity)')\n", "plt.show()\n", "\n", "print('Intercept: {}\\n'\n", " 'Slope: {}\\n'\n", " 'Correlation: {}\\n'\n", " 'R-squared: {}\\n'\n", " 'p-value: {}\\n'\n", " 't-statistic: {}'\n", " .format(intercept, slope, r_value, r_value**2, p_value, slope/std_err))" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Evaluation\n", "We want to determine whether sentiment and price have a positive correlation. Lets test this against the null hypothesis that their correlation coefficient is zero.\n", "\n", "Let us declare our null and alternative hypotheses:\n", "\n", "* $\\text{Hypothesis H}_0: \\rho = 0$\n", "* $\\text{Hypothesis H}_A: \\rho > 0$\n", "\n", "From our above regression, we have: $\\text{p-value} = Pr(T > t) = 0.0882 \\text{.}$\n", "\n", "Testing with a confidence level of 0.05, we fall outside the critical region. Thus we **fail to reject** our null hypothesis.\n", "\n", "## How to Improve\n", "To see the results we were expecting, we could run our inference on a dataset from a wider time interval, thereby reducing our standard error and p-value. \n", "\n", "In the sentiment analysis phase, we could additionally tune our classifier by training with domain-specific language and vocabulary. In the case that the keyword for a currency may return Tweets totally unrelated to what we intend (such as the case with *Dash* or *Neo*), we end up factoring in junk data into our average polarity scores. Filtering out these results would inprove the validity and accuracy of our inference." ] } ], "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": "3.6.4" } }, "nbformat": 4, "nbformat_minor": 2 }