The code runs free of errors but always gives Nan correlation. i dont understand what the problem is$,$ i tried every solution but keeps producing same output. i have attached the $4$ datasets i have to make the code and create a correlation for the four hupothesis and even attach the result. my data frame is always empty no matter how i edit the code. there is no null values or string values in my data set as well $.$ help me and give me a code free of error and give me right correlation. removing rows, merging, eberything is done but still gives Nan correlation. give the correct code, no need for explanation if the code still gives Nan output.

import pandas as pd

import numpy as np

from scipy.stats import pearsonr

import matplotlib.pyplot as plt

# Read data files

house$\_$prices$\_$df $=$ pd$.$read$\_$excel$("$MeanHousePricesClean$-1.$xlsx$")$

crime$\_$df $=$ pd$.$read$\_$excel$("$CrimeClean$-1-1.$xlsx$")$

population$\_$df $=$ pd$.$read$\_$excel$("$PopulationClean$.$xlsx$")$

area$\_$df $=$ pd$.$read$\_$excel$("$SuburbAreas$-1.$xlsx$",$ header$=$None$)$

# Step B: Clean and prepare data

def prepare$\_$data$($df$,$ columns$)$:

df $=$ df$.$dropna$($subset$=$columns$)$ # Remove rows with missing values in key columns

return df

# Rename columns for consistency

house$\_$prices$\_$df $=$ house$\_$prices$\_$df$.$rename$($columns$=\{\text{'}$Year$\text{'}$: 'year'$\})$

crime$\_$df $=$ crime$\_$df$.$rename$($columns$=\{\text{'}$Year$\text{'}$: 'year', 'Crime rate per $100,000$ population': 'crime$\_$rate',

'Local Government Area': 'local$\_$government$\_$area'$\})$

population$\_$df $=$ population$\_$df$.$rename$($columns$=\{\text{'}$Year$\text{'}$: 'year'$\})$

area$\_$df $=$ area$\_$df$.$rename$($columns$=\{0$: 'local$\_$government$\_$area', $1$: 'area'$\})$

# Clean the area DataFrame to remove non$-$relevant rows

area$\_$df $=$ area$\_$df$[$area$\_$df$[\text{'}$local$\_$government$\_$area'$]!=$ 'Property'$]$

# Step C: Analysis functions

def analyze$\_$correlation$($df$,$ col$1,$ col$2)$:

df $=$ df$.$dropna$($subset$=[$col$1,$ col$2])$

if len$($df$)<mtext\; id="EditorElement\_1544779"></mtext><mn\; id="EditorElement\_1544780">2</mn>$:

return np$.$nan

correlation, $\_=$ pearsonr$($df$[$col$1],$ df$[$col$2])$

return correlation

# Reshape data to long format

house$\_$prices$\_$long $=$ pd$.$melt$($house$\_$prices$\_$df$,$ id$\_$vars$=[\text{'}$year$\text{'}],$ var$\_$name$=$'local$\_$government$\_$area', value$\_$name$=$'mean$\_$house$\_$price'$)$

population$\_$long $=$ pd$.$melt$($population$\_$df$,$ id$\_$vars$=[\text{'}$year$\text{'}],$ var$\_$name$=$'local$\_$government$\_$area', value$\_$name$=$'population'$)$

# Merge the datasets on 'year' and 'local$\_$government$\_$area'

merged$\_$df $=$ pd$.$merge$($crime$\_$df$,$ house$\_$prices$\_$long, on$=[\text{'}$year$\text{'},$ 'local$\_$government$\_$area'$],$ how$=$'inner'$)$

merged$\_$df $=$ pd$.$merge$($merged$\_$df$,$ population$\_$long, on$=[\text{'}$year$\text{'},$ 'local$\_$government$\_$area'$],$ how$=$'inner'$)$

merged$\_$df $=$ pd$.$merge$($merged$\_$df$,$ area$\_$long, on$=$'local$\_$government$\_$area', how$=$'inner'$)$

# Calculate population density

merged$\_$df$[\text{'}$population$\_$density'$]=$ merged$\_$df$[\text{'}$population$\text{'}]/$ merged$\_$df$[\text{'}$area$\text{'}]$

# Step D: Prepare the data by cleaning

merged$\_$df $=$ prepare$\_$data$($merged$\_$df$,[\text{'}$mean$\_$house$\_$price', 'crime$\_$rate', 'population$\_$density'$])$

# Step E: Perform correlation analysis

house$\_$price$\_$population$\_$corr $=$ analyze$\_$correlation$($merged$\_$df$,$ 'mean$\_$house$\_$price', 'population$\_$density'$)$

crime$\_$house$\_$price$\_$corr $=$ analyze$\_$correlation$($merged$\_$df$,$ 'crime$\_$rate', 'mean$\_$house$\_$price'$)$

crime$\_$population$\_$density$\_$corr $=$ analyze$\_$correlation$($merged$\_$df$,$ 'crime$\_$rate', 'population$\_$density'$)$

# Step F: Print the results

print$($f$"$Correlation between house prices and population density: $\{$house$\_$price$\_$population$\_$corr$\}")$

print$($f$"$Correlation between crime rate and house prices: $\{$crime$\_$house$\_$price$\_$corr$\}")$

print$($f$"$Correlation between crime rate and population density: $\{$crime$\_$population$\_$density$\_$corr$\}")$

# Plotting for visual analysis

plt$.$figure$($figsize$=(10,6))$

plt$.$scatter$($merged$\_$df$[\text{'}$population$\_$density'$],$ merged$\_$df$[\text{'}$mean$\_$house$\_$price'$])$

plt$.$title$(\text{'}$House Price vs Population Density'$)$

plt$.$xlabel$(\text{'}$Population Density $($people per square km$)\text{'})$

plt$.$ylabel$(\text{'}$Mean House Price'$)$

plt$.$grid$($True$)$

plt$.$show$()$

plt$.$figure$($figsize$=(10,6))$

plt$.$scatter$($merged$\_$df$[\text{'}$mean$\_$house$\_$price'$],$ merged$\_$df$[\text{'}$crime$\_$rate'$])$

plt$.$title$(\text{'}$Crime Rate vs House Price'$)$

plt$.$xlabel$(\text{'}$Mean House Price'$)$

plt$.$ylabel$(\text{'}$Crime Rate $($per $100,000$ population$)\text{'})$

plt$.$grid$($True$)$

plt$.$show$()$

plt$.$figure$($figsize$=(10,6))$

plt$.$scatter$($merged$\_$df$[\text{'}$population$\_$density'$],$ merged$\_$df$[\text{'}$crime$\_$rate'$])$

plt$.$title$(\text{'}$Crime Rate vs Population Density'$)$

plt$.$xlabel$(\text{'}$Population Density $($people per square km$)\text{'})$

plt$.$ylabel$(\text{'}$Crime Rate $($per $100,000$ population$)\text{'})$

plt$.$grid$($True$)$

plt$.$show$()$

output the above code gives: Correlation between house prices and population density: nan

Correlation between crime rate and house prices: nan

Correlation between crime rate and population density: nan

problem: Empty DataFrame

Columns: $[$Incidents recorded, crime$\_$rate, mean$\_$house$\_$price, year, population, local$\_$government$\_$area, area, population$\_$density$]$

Index: $[]$