diff --git a/lab-hypothesis-testing.ipynb b/lab-hypothesis-testing.ipynb index 0cc26d5..8ea8912 100644 --- a/lab-hypothesis-testing.ipynb +++ b/lab-hypothesis-testing.ipynb @@ -38,7 +38,7 @@ }, { "cell_type": "code", - "execution_count": 1, + "execution_count": 4, "metadata": {}, "outputs": [], "source": [ @@ -51,11 +51,782 @@ }, { "cell_type": "code", - "execution_count": 3, + "execution_count": 5, "metadata": {}, "outputs": [ { "data": { + "application/vnd.microsoft.datawrangler.viewer.v0+json": { + "columns": [ + { + "name": "index", + "rawType": "int64", + "type": "integer" + }, + { + "name": "Name", + "rawType": "object", + "type": "unknown" + }, + { + "name": "Type 1", + "rawType": "object", + "type": "string" + }, + { + "name": "Type 2", + "rawType": "object", + "type": "unknown" + }, + { + "name": "HP", + "rawType": "int64", + "type": "integer" + }, + { + "name": "Attack", + "rawType": "int64", + "type": "integer" + }, + { + "name": "Defense", + "rawType": "int64", + "type": "integer" + }, + { + "name": "Sp. Atk", + "rawType": "int64", + "type": "integer" + }, + { + "name": "Sp. Def", + "rawType": "int64", + "type": "integer" + }, + { + "name": "Speed", + "rawType": "int64", + "type": "integer" + }, + { + "name": "Generation", + "rawType": "int64", + "type": "integer" + }, + { + "name": "Legendary", + "rawType": "bool", + "type": "boolean" + } + ], + "ref": "fc83ee43-3faa-4948-8c69-72701277fa60", + "rows": [ + [ + "0", + "Bulbasaur", + "Grass", + "Poison", + "45", + "49", + "49", + "65", + "65", + "45", + "1", + "False" + ], + [ + "1", + "Ivysaur", + "Grass", + "Poison", + "60", + "62", + "63", + "80", + "80", + "60", + "1", + "False" + ], + [ + "2", + "Venusaur", + "Grass", + "Poison", + "80", + "82", + "83", + "100", + "100", + "80", + "1", + "False" + ], + [ + "3", + "Mega Venusaur", + "Grass", + "Poison", + "80", + "100", + "123", + "122", + "120", + "80", + "1", + "False" + ], + [ + "4", + "Charmander", + "Fire", + null, + "39", + "52", + "43", + "60", + "50", + "65", + "1", + "False" + ], + [ + "5", + "Charmeleon", + "Fire", + null, + "58", + "64", + "58", + "80", + "65", + "80", + "1", + "False" + ], + [ + "6", + "Charizard", + "Fire", + "Flying", + "78", + "84", + "78", + "109", + "85", + "100", + "1", + "False" + ], + [ + "7", + "Mega Charizard X", + "Fire", + "Dragon", + "78", + "130", + "111", + "130", + "85", + "100", + "1", + "False" + ], + [ + "8", + "Mega Charizard Y", + "Fire", + "Flying", + "78", + "104", + "78", + "159", + "115", + "100", + "1", + "False" + ], + [ + "9", + "Squirtle", + "Water", + null, + "44", + "48", + "65", + "50", + "64", + "43", + "1", + "False" + ], + [ + "10", + "Wartortle", + "Water", + null, + "59", + "63", + "80", + "65", + "80", + "58", + "1", + "False" + ], + [ + "11", + "Blastoise", + "Water", + null, + "79", + "83", + "100", + "85", + "105", + "78", + "1", + "False" + ], + [ + "12", + "Mega Blastoise", + "Water", + null, + "79", + "103", + "120", + "135", + "115", + "78", + "1", + "False" + ], + [ + "13", + "Caterpie", + "Bug", + null, + "45", + "30", + "35", + "20", + "20", + "45", + "1", + "False" + ], + [ + "14", + "Metapod", + "Bug", + null, + "50", + "20", + "55", + "25", + "25", + "30", + "1", + "False" + ], + [ + "15", + "Butterfree", + "Bug", + "Flying", + "60", + "45", + "50", + "90", + "80", + "70", + "1", + "False" + ], + [ + "16", + "Weedle", + "Bug", + "Poison", + "40", + "35", + "30", + "20", + "20", + "50", + "1", + "False" + ], + [ + "17", + "Kakuna", + "Bug", + "Poison", + "45", + "25", + "50", + "25", + "25", + "35", + "1", + "False" + ], + [ + "18", + "Beedrill", + "Bug", + "Poison", + "65", + "90", + "40", + "45", + "80", + "75", + "1", + "False" + ], + [ + "19", + "Mega Beedrill", + "Bug", + "Poison", + "65", + "150", + "40", + "15", + "80", + "145", + "1", + "False" + ], + [ + "20", + "Pidgey", + "Normal", + "Flying", + "40", + "45", + "40", + "35", + "35", + "56", + "1", + "False" + ], + [ + "21", + "Pidgeotto", + "Normal", + "Flying", + "63", + "60", + "55", + "50", + "50", + "71", + "1", + "False" + ], + [ + "22", + "Pidgeot", + "Normal", + "Flying", + "83", + "80", + "75", + "70", + "70", + "101", + "1", + "False" + ], + [ + "23", + "Mega Pidgeot", + "Normal", + "Flying", + "83", + "80", + "80", + "135", + "80", + "121", + "1", + "False" + ], + [ + "24", + "Rattata", + "Normal", + null, + "30", + "56", + "35", + "25", + "35", + "72", + "1", + "False" + ], + [ + "25", + "Raticate", + "Normal", + null, + "55", + "81", + "60", + "50", + "70", + "97", + "1", + "False" + ], + [ + "26", + "Spearow", + "Normal", + "Flying", + "40", + "60", + "30", + "31", + "31", + "70", + "1", + "False" + ], + [ + "27", + "Fearow", + "Normal", + "Flying", + "65", + "90", + "65", + "61", + "61", + "100", + "1", + "False" + ], + [ + "28", + "Ekans", + "Poison", + null, + "35", + "60", + "44", + "40", + "54", + "55", + "1", + "False" + ], + [ + "29", + "Arbok", + "Poison", + null, + "60", + "85", + "69", + "65", + "79", + "80", + "1", + "False" + ], + [ + "30", + "Pikachu", + "Electric", + null, + "35", + "55", + "40", + "50", + "50", + "90", + "1", + "False" + ], + [ + "31", + "Raichu", + "Electric", + null, + "60", + "90", + "55", + "90", + "80", + "110", + "1", + "False" + ], + [ + "32", + "Sandshrew", + "Ground", + null, + "50", + "75", + "85", + "20", + "30", + "40", + "1", + "False" + ], + [ + "33", + "Sandslash", + "Ground", + null, + "75", + "100", + "110", + "45", + "55", + "65", + "1", + "False" + ], + [ + "34", + "Nidoran♀", + "Poison", + null, + "55", + "47", + "52", + "40", + "40", + "41", + "1", + "False" + ], + [ + "35", + "Nidorina", + "Poison", + null, + "70", + "62", + "67", + "55", + "55", + "56", + "1", + "False" + ], + [ + "36", + "Nidoqueen", + "Poison", + "Ground", + "90", + "92", + "87", + "75", + "85", + "76", + "1", + "False" + ], + [ + "37", + "Nidoran♂", + "Poison", + null, + "46", + "57", + "40", + "40", + "40", + "50", + "1", + "False" + ], + [ + "38", + "Nidorino", + "Poison", + null, + "61", + "72", + "57", + "55", + "55", + "65", + "1", + "False" + ], + [ + "39", + "Nidoking", + "Poison", + "Ground", + "81", + "102", + "77", + "85", + "75", + "85", + "1", + "False" + ], + [ + "40", + "Clefairy", + "Fairy", + null, + "70", + "45", + "48", + "60", + "65", + "35", + "1", + "False" + ], + [ + "41", + "Clefable", + "Fairy", + null, + "95", + "70", + "73", + "95", + "90", + "60", + "1", + "False" + ], + [ + "42", + "Vulpix", + "Fire", + null, + "38", + "41", + "40", + "50", + "65", + "65", + "1", + "False" + ], + [ + "43", + "Ninetales", + "Fire", + null, + "73", + "76", + "75", + "81", + "100", + "100", + "1", + "False" + ], + [ + "44", + "Jigglypuff", + "Normal", + "Fairy", + "115", + "45", + "20", + "45", + "25", + "20", + "1", + "False" + ], + [ + "45", + "Wigglytuff", + "Normal", + "Fairy", + "140", + "70", + "45", + "85", + "50", + "45", + "1", + "False" + ], + [ + "46", + "Zubat", + "Poison", + "Flying", + "40", + "45", + "35", + "30", + "40", + "55", + "1", + "False" + ], + [ + "47", + "Golbat", + "Poison", + "Flying", + "75", + "80", + "70", + "65", + "75", + "90", + "1", + "False" + ], + [ + "48", + "Oddish", + "Grass", + "Poison", + "45", + "50", + "55", + "75", + "65", + "30", + "1", + "False" + ], + [ + "49", + "Gloom", + "Grass", + "Poison", + "60", + "65", + "70", + "85", + "75", + "40", + "1", + "False" + ] + ], + "shape": { + "columns": 11, + "rows": 800 + } + }, "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", + "
longitudelatitudehousing_median_agetotal_roomstotal_bedroomspopulationhouseholdsmedian_incomemedian_house_valueproximity
0-114.3134.1915.05612.01283.01015.0472.01.493666900.0far
1-114.4734.4019.07650.01901.01129.0463.01.820080100.0far
2-114.5633.6917.0720.0174.0333.0117.01.650985700.0far
3-114.5733.6414.01501.0337.0515.0226.03.191773400.0far
4-114.5733.5720.01454.0326.0624.0262.01.925065500.0far
.................................
16995-124.2640.5852.02217.0394.0907.0369.02.3571111400.0far
16996-124.2740.6936.02349.0528.01194.0465.02.517979000.0far
16997-124.3041.8417.02677.0531.01244.0456.03.0313103600.0far
16998-124.3041.8019.02672.0552.01298.0478.01.979785800.0far
16999-124.3540.5452.01820.0300.0806.0270.03.014794600.0far
\n", + "

17000 rows × 10 columns

\n", + "
" + ], + "text/plain": [ + " longitude latitude housing_median_age total_rooms total_bedrooms \\\n", + "0 -114.31 34.19 15.0 5612.0 1283.0 \n", + "1 -114.47 34.40 19.0 7650.0 1901.0 \n", + "2 -114.56 33.69 17.0 720.0 174.0 \n", + "3 -114.57 33.64 14.0 1501.0 337.0 \n", + "4 -114.57 33.57 20.0 1454.0 326.0 \n", + "... ... ... ... ... ... \n", + "16995 -124.26 40.58 52.0 2217.0 394.0 \n", + "16996 -124.27 40.69 36.0 2349.0 528.0 \n", + "16997 -124.30 41.84 17.0 2677.0 531.0 \n", + "16998 -124.30 41.80 19.0 2672.0 552.0 \n", + "16999 -124.35 40.54 52.0 1820.0 300.0 \n", + "\n", + " population households median_income median_house_value proximity \n", + "0 1015.0 472.0 1.4936 66900.0 far \n", + "1 1129.0 463.0 1.8200 80100.0 far \n", + "2 333.0 117.0 1.6509 85700.0 far \n", + "3 515.0 226.0 3.1917 73400.0 far \n", + "4 624.0 262.0 1.9250 65500.0 far \n", + "... ... ... ... ... ... \n", + "16995 907.0 369.0 2.3571 111400.0 far \n", + "16996 1194.0 465.0 2.5179 79000.0 far \n", + "16997 1244.0 456.0 3.0313 103600.0 far \n", + "16998 1298.0 478.0 1.9797 85800.0 far \n", + "16999 806.0 270.0 3.0147 94600.0 far \n", + "\n", + "[17000 rows x 10 columns]" + ] + }, + "execution_count": 42, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "# Define function for calculating euclidean distance, and measuring how far it is from school and hospital, and return close or far, for close distance it should be lower than 0.50\n", + "\n", + "import numpy as np\n", + "from scipy.spatial.distance import euclidean\n", + "\n", + "\n", + "def classify_proximity(house_coords):\n", + " school_coords = (-118, 34)\n", + " hospital_coords = (-122, 38)\n", + " school_dist = euclidean(house_coords, school_coords)\n", + " hospital_dist = euclidean(house_coords, hospital_coords)\n", + " if (school_dist < 0.5) or (hospital_dist < 0.5):\n", + " return \"close\"\n", + " else:\n", + " return \"far\"\n", + "\n", + "\n", + "# Apply to DataFrame\n", + "for index, row in df.iterrows():\n", + " # Extract house coordinates from the table\n", + " house_coords = (row[\"longitude\"], row[\"latitude\"])\n", + " proximity = classify_proximity(house_coords)\n", + " df.at[index, \"proximity\"] = proximity\n", + "\n", + "\n", + "df" + ] }, { "cell_type": "code", - "execution_count": null, + "execution_count": 43, "metadata": {}, "outputs": [], - "source": [] + "source": [ + "# H0: close houses prices > far houses prices\n", + "# H1: close houses prices <= far houses prices\n", + "\n", + "# Significance level = 0.05\n", + "\n", + "close_prices = df[df[\"proximity\"]==\"close\"][\"median_house_value\"]\n", + "far_prices = df[df[\"proximity\"]==\"far\"][\"median_house_value\"]\n" + ] + }, + { + "cell_type": "code", + "execution_count": 45, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "TtestResult(statistic=np.float64(26.871988399009897), pvalue=np.float64(1.0), df=np.float64(16998.0))" + ] + }, + "execution_count": 45, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "# The p-value is higher than 0.5, we fail to reject the null hypothesis.\n", + "# Therefore, houses close to school and hospital have higher median house values than those far from both.\n", + "\n", + "st.ttest_ind(close_prices, far_prices, alternative=\"less\")" + ] } ], "metadata": { "kernelspec": { - "display_name": "Python 3", + "display_name": "base", "language": "python", "name": "python3" }, @@ -512,7 +2581,7 @@ "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", - "version": "3.10.9" + "version": "3.13.5" } }, "nbformat": 4,