{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "# Test3: (40 marks)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "# Your ID: 6304640318\n" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "In this excercise, we aimed at the case of customers default payments in Taiwan and compares the predictive accuracy of probability of default among data mining methods. From the perspective of risk management, the result of predictive accuracy of the estimated probability of default will be more valuable than the binary result of classification - credible or not credible clients. With the real probability of default as the response variable (Y), and the predictive probability of default as the independent variable (X), this research employed a binary variable, default payment (Yes = 1, No = 0), as the response variable. This study reviewed the literature and used the following 23 variables as explanatory variables: \\\\\n", "\n", "X1: Amount of the given credit (NT dollar): it includes both the individual consumer credit and his/her family (supplementary) credit. \\\\\n", "\n", "X2: Gender (1 = male; 2 = female). \\\\\n", "\n", "X3: Education (1 = graduate school; 2 = university; 3 = high school; 4 = others). \\\\\n", "\n", "X4: Marital status (1 = married; 2 = single; 3 = others). \\\\\n", "\n", "X5: Age (year). \\\\\n", "\n", "X6 - X11: History of past payment. We tracked the past monthly payment records (from April to September, 2005) as follows: X6 = the repayment status in September, 2005; X7 = the repayment status in August, 2005; . . .;X11 = the repayment status in April, 2005. The measurement scale for the repayment status is: -1 = pay duly; 1 = payment delay for one month; 2 = payment delay for two months; . . .; 8 = payment delay for eight months; 9 = payment delay for nine months and above. \\\\\n", "\n", "X12-X17: Amount of bill statement (NT dollar). X12 = amount of bill statement in September, 2005; X13 = amount of bill statement in August, 2005; . . .; X17 = amount of bill statement in April, 2005. \\\\\n", "\n", "X18-X23: Amount of previous payment (NT dollar). X18 = amount paid in September, 2005; X19 = amount paid in August, 2005; . . .;X23 = amount paid in April, 2005. \\\\\n" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "# Learning the Logistic Regression, KNN, in Python and Scikit-Learn \n", "\n", "There are 4 sub-question:\n", "\n", "1.1 Create the table to report the proportion of case of customers default payments in Taiwan separated by gender, education, and marital status. What is/are the interesting result/s you can draw from this table?[10 Points].\\\\" ] }, { "cell_type": "code", "execution_count": 22, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "no display found. Using non-interactive Agg backend\n" ] } ], "source": [ "import numpy as np\n", "import pandas as pd\n", "import seaborn as sns\n", "from sklearn import (\n", " linear_model, metrics, pipeline, preprocessing, model_selection)\n", "%matplotlib inline\n", "\n", "from pathlib import Path\n", "\n", "import pandas as pd\n", "from sklearn.model_selection import train_test_split\n", "from sklearn.neural_network import MLPClassifier\n", "\n", "from sklearn import metrics #evaluation model.\n", "from dmba import classificationSummary" ] }, { "cell_type": "code", "execution_count": 52, "metadata": {}, "outputs": [], "source": [ "df = pd.read_excel(\"default.xls\")" ] }, { "cell_type": "code", "execution_count": 53, "metadata": { "scrolled": true }, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
Unnamed: 0X1X2X3X4X5X6X7X8X9...X15X16X17X18X19X20X21X22X23Y
01200002212422-1-1...000068900001
1212000022226-1200...3272345532610100010001000020001
2390000222340000...1433114948155491518150010001000100050000
3450000221370000...2831428959295472000201912001100106910000
455000012157-10-10...2094019146191312000366811000090006896790
..................................................................
2999529996220000131390000...88004312371598085002000050033047500010000
299962999715000013243-1-1-1-1...897951900183735268998129000
29997299983000012237432-1...20878205821935700220004200200031001
299982999980000131411-100...527741185548944859003409117819265296418041
299993000050000121460000...3653532428153132078180014301000100010001
\n", "

30000 rows × 25 columns

\n", "
" ], "text/plain": [ " Unnamed: 0 X1 X2 X3 X4 X5 X6 X7 X8 X9 ... X15 X16 \\\n", "0 1 20000 2 2 1 24 2 2 -1 -1 ... 0 0 \n", "1 2 120000 2 2 2 26 -1 2 0 0 ... 3272 3455 \n", "2 3 90000 2 2 2 34 0 0 0 0 ... 14331 14948 \n", "3 4 50000 2 2 1 37 0 0 0 0 ... 28314 28959 \n", "4 5 50000 1 2 1 57 -1 0 -1 0 ... 20940 19146 \n", "... ... ... .. .. .. .. .. .. .. .. ... ... ... \n", "29995 29996 220000 1 3 1 39 0 0 0 0 ... 88004 31237 \n", "29996 29997 150000 1 3 2 43 -1 -1 -1 -1 ... 8979 5190 \n", "29997 29998 30000 1 2 2 37 4 3 2 -1 ... 20878 20582 \n", "29998 29999 80000 1 3 1 41 1 -1 0 0 ... 52774 11855 \n", "29999 30000 50000 1 2 1 46 0 0 0 0 ... 36535 32428 \n", "\n", " X17 X18 X19 X20 X21 X22 X23 Y \n", "0 0 0 689 0 0 0 0 1 \n", "1 3261 0 1000 1000 1000 0 2000 1 \n", "2 15549 1518 1500 1000 1000 1000 5000 0 \n", "3 29547 2000 2019 1200 1100 1069 1000 0 \n", "4 19131 2000 36681 10000 9000 689 679 0 \n", "... ... ... ... ... ... ... ... .. \n", "29995 15980 8500 20000 5003 3047 5000 1000 0 \n", "29996 0 1837 3526 8998 129 0 0 0 \n", "29997 19357 0 0 22000 4200 2000 3100 1 \n", "29998 48944 85900 3409 1178 1926 52964 1804 1 \n", "29999 15313 2078 1800 1430 1000 1000 1000 1 \n", "\n", "[30000 rows x 25 columns]" ] }, "execution_count": 53, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df" ] }, { "cell_type": "code", "execution_count": 54, "metadata": {}, "outputs": [], "source": [ "OM1 = df[[\"X1\",\"X2\",\"X3\",'X4']]" ] }, { "cell_type": "code", "execution_count": 55, "metadata": { "scrolled": true }, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
X1X2X3X4
020000221
1120000222
290000222
350000221
450000121
...............
29995220000131
29996150000132
2999730000122
2999880000131
2999950000121
\n", "

30000 rows × 4 columns

\n", "
" ], "text/plain": [ " X1 X2 X3 X4\n", "0 20000 2 2 1\n", "1 120000 2 2 2\n", "2 90000 2 2 2\n", "3 50000 2 2 1\n", "4 50000 1 2 1\n", "... ... .. .. ..\n", "29995 220000 1 3 1\n", "29996 150000 1 3 2\n", "29997 30000 1 2 2\n", "29998 80000 1 3 1\n", "29999 50000 1 2 1\n", "\n", "[30000 rows x 4 columns]" ] }, "execution_count": 55, "metadata": {}, "output_type": "execute_result" } ], "source": [ "OM1" ] }, { "cell_type": "code", "execution_count": 56, "metadata": { "scrolled": true }, "outputs": [], "source": [ "y = df[['Y']]\n" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Then, partition the data into training and test sets using 80 $\\%$ to be the train data. The model will be fit to the training data and evaluated on the test set. \n", "\n", "1.2 Use the Logistic regression to train the model with all features. Then, use the test data to conduct the confusion matrix. Interpret the results carefully [10 Points]." ] }, { "cell_type": "code", "execution_count": 57, "metadata": {}, "outputs": [], "source": [ "from sklearn.model_selection import train_test_split" ] }, { "cell_type": "code", "execution_count": 58, "metadata": {}, "outputs": [], "source": [ "from sklearn import (\n", " linear_model, metrics, pipeline, preprocessing, model_selection\n", ")" ] }, { "cell_type": "code", "execution_count": 59, "metadata": {}, "outputs": [], "source": [ "X_train, X_test, y_train, y_test = model_selection.train_test_split(OM1,y,test_size=0.20)" ] }, { "cell_type": "code", "execution_count": 60, "metadata": { "scrolled": false }, "outputs": [ { "data": { "text/plain": [ "LogisticRegression()" ] }, "execution_count": 60, "metadata": {}, "output_type": "execute_result" } ], "source": [ "linear_model.LogisticRegression()" ] }, { "cell_type": "code", "execution_count": 61, "metadata": {}, "outputs": [], "source": [ "logistic_model=linear_model.LogisticRegression()" ] }, { "cell_type": "code", "execution_count": 62, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Train accuracy: 0.78\n", "Test accuracy: 0.78\n" ] }, { "name": "stderr", "output_type": "stream", "text": [ "/opt/anaconda3/lib/python3.9/site-packages/sklearn/utils/validation.py:993: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", " y = column_or_1d(y, warn=True)\n" ] } ], "source": [ "from sklearn.linear_model import LogisticRegression\n", "clf = LogisticRegression(max_iter=2500)\n", "\n", "# Train the model\n", "clf.fit(X_train, y_train)\n", "\n", "# Make predictions\n", "y_pred = clf.predict(X_test) # Predictions\n", "y_true = y_test # True values\n", "\n", "# Measure accuracy\n", "from sklearn.metrics import accuracy_score\n", "import numpy as np\n", "print(\"Train accuracy:\", np.round(accuracy_score(y_train, \n", " clf.predict(X_train)), 2))\n", "print(\"Test accuracy:\", np.round(accuracy_score(y_true, y_pred), 2))\n", "\n", "# Make the confusion matrix\n", "from sklearn.metrics import confusion_matrix\n", "cf_matrix = confusion_matrix(y_true, y_pred)" ] }, { "cell_type": "code", "execution_count": 64, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ " precision recall f1-score support\n", "\n", " default 0.78 1.00 0.87 4660\n", " not default 0.00 0.00 0.00 1340\n", "\n", " accuracy 0.78 6000\n", " macro avg 0.39 0.50 0.44 6000\n", "weighted avg 0.60 0.78 0.68 6000\n", "\n" ] }, { "name": "stderr", "output_type": "stream", "text": [ "/opt/anaconda3/lib/python3.9/site-packages/sklearn/metrics/_classification.py:1318: UndefinedMetricWarning: Precision and F-score are ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n", " _warn_prf(average, modifier, msg_start, len(result))\n", "/opt/anaconda3/lib/python3.9/site-packages/sklearn/metrics/_classification.py:1318: UndefinedMetricWarning: Precision and F-score are ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n", " _warn_prf(average, modifier, msg_start, len(result))\n", "/opt/anaconda3/lib/python3.9/site-packages/sklearn/metrics/_classification.py:1318: UndefinedMetricWarning: Precision and F-score are ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n", " _warn_prf(average, modifier, msg_start, len(result))\n" ] } ], "source": [ "report = metrics.classification_report(\n", " y_test, clf.predict(X_test),\n", " target_names=[\"default\",\"not default\"]\n", ")\n", "print(report)" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [] }, { "cell_type": "markdown", "metadata": {}, "source": [ "1.3 Use the KNN to train the model with all features (you should convert the features to be the standarized variables first). Then, use the test data to conduct the confusion matrix. Interpret the results carefully [10 Points]." ] }, { "cell_type": "code", "execution_count": 32, "metadata": {}, "outputs": [], "source": [ "from sklearn.neighbors import KNeighborsClassifier\n", "from sklearn.tree import DecisionTreeClassifier\n", "from sklearn.ensemble import RandomForestClassifier\n", "from sklearn.linear_model import LogisticRegression\n", "\n", "classifier = KNeighborsClassifier(n_neighbors=9)\n", "from sklearn.model_selection import KFold\n", "import numpy as np" ] }, { "cell_type": "code", "execution_count": 33, "metadata": {}, "outputs": [], "source": [ "kf=KFold(n_splits=5, shuffle= True)" ] }, { "cell_type": "code", "execution_count": 37, "metadata": {}, "outputs": [], "source": [ "from sklearn.neighbors import KNeighborsClassifier" ] }, { "cell_type": "code", "execution_count": 35, "metadata": {}, "outputs": [], "source": [ "from sklearn.preprocessing import StandardScaler\n", "sc = StandardScaler()\n", "X_scaled = sc.fit_transform(OM1)" ] }, { "cell_type": "code", "execution_count": 38, "metadata": {}, "outputs": [], "source": [ "knn = KNeighborsClassifier(n_neighbors=7)" ] }, { "cell_type": "code", "execution_count": 42, "metadata": { "scrolled": true }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "0.7583333333333333\n" ] }, { "name": "stderr", "output_type": "stream", "text": [ "/opt/anaconda3/lib/python3.9/site-packages/sklearn/neighbors/_classification.py:198: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples,), for example using ravel().\n", " return self._fit(X, y)\n" ] } ], "source": [ "X_train, X_test, y_train, y_test = train_test_split(\n", " X_scaled, y, test_size = 0.2)\n", "knn.fit(X_train, y_train)\n", "y_pred= knn.predict(X_test)\n", "from sklearn.metrics import accuracy_score\n", "\n", "# printing accuracy\n", "print(accuracy_score(y_test,y_pred))" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "y_pred= knn.predict(X_test)" ] }, { "cell_type": "code", "execution_count": 44, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ " precision recall f1-score support\n", "\n", " 0 0.78 0.95 0.86 4680\n", " 1 0.30 0.08 0.12 1320\n", "\n", " accuracy 0.76 6000\n", " macro avg 0.54 0.51 0.49 6000\n", "weighted avg 0.68 0.76 0.70 6000\n", "\n" ] } ], "source": [ "from sklearn.metrics import accuracy_score\n", "accuracy_score(y_test,y_pred)\n", "\n", "# finding the whole report\n", "from sklearn.metrics import classification_report\n", "print(classification_report(y_test, y_pred))\n" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ " 1.4 Compare the predictive performance of two models. Which one you select? And Why? [10 Points]" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "# with the normal logistic regression we can get the higher accuracy rate \n", "# and the recall rate that provid use with the high recall rate that knn model but with the \n", "#logistic regression we can not predict the result for default but with the knn we can predict with the\n", "#defualt part give us the score but with the low result" ] } ], "metadata": { "kernelspec": { "display_name": "Python 3 (ipykernel)", "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": "3.9.12" } }, "nbformat": 4, "nbformat_minor": 4 }