import numpy as np

import pandas as pd

import matplotlib.pyplot as plt

​

import seaborn as sns

import statsmodels.api as sm

df = pd.read_csv("D:/Users/User/Documents/Application Harvard University/Harvard University 2025/MCSO Fall 2025/CSO 707 Quantitative Science and Clinical Research in Healthcare Service Operations/ED_LOS.csv")

df.info()

df.head()

df["Network"].unique()

## Perform Linear Regression

from sklearn.linear_model import LinearRegression

​

x = df["ALOS_Avg"]

y = df["EDOS_Avg"]

 plt.plot(x, y, 'o', zorder = 4, label="df")

x = df["ALOS_Avg"]

y = df["EDOS_Avg"]

​

# 2. Set the style for better visualization (optional)

sns.set_style("whitegrid")

​

# 3. Create the plot

# 'ci=95' adds a 95% confidence interval band around the regression line.

plt.figure(figsize=(8, 5))

sns.regplot(x= df["ALOS_Avg"], y=df["EDOS_Avg"], data=df, ci=95, scatter_kws={'color': 'darkblue'}, line_kws={'color': 'red'})

​

# 4. Add titles and labels

plt.title('Regression Plot of ALOS vs EDOS')

plt.xlabel('ALOS')

plt.ylabel('EDOS')

​

# 5. Display the plot

plt.show()

​

sns.lmplot(data = df, x="ALOS_Avg", y="EDOS_Avg")

​

# Visualizing the Relationship between ALOS and EDOS for each category of Network

sns.lmplot(data = df, x='ALOS_Avg', y='EDOS_Avg', hue='Network')

​

plt.show()

​

df["Network"].nunique()

# Distribution of ALOS

# Plotting the distribution

plt.figure(figsize=(8, 5))

sns.histplot(df['ALOS_Avg'], kde=True) # kde=True adds a Kernel Density Estimate line

plt.title('Distribution of ALOS')

plt.xlabel('Avg ALOS')

plt.ylabel('Frequency/Density')

​

ALOS_mean = df['ALOS_Avg'].mean()

ALOS_std = df['ALOS_Avg'].std()

​

# Optional: Add vertical lines for mean and +/- 1 standard deviation

plt.axvline(ALOS_mean, color='r', linestyle='--', label=f'Mean: {ALOS_mean:.2f}')

plt.axvline(ALOS_mean - ALOS_std, color='g', linestyle=':', label=f'-1 Std Dev: {ALOS_mean - ALOS_std:.2f}')

plt.axvline(ALOS_mean + ALOS_std, color='g', linestyle=':', label=f'+1 Std Dev: {ALOS_mean + ALOS_std:.2f}')

plt.legend()

# Distribution of ALOS

# Plotting the distribution

plt.figure(figsize=(8, 5))

sns.histplot(df['EDOS_Avg'], kde=True) # kde=True adds a Kernel Density Estimate line

plt.title('Distribution of EDOS')

plt.xlabel('Avg EDOS')

plt.ylabel('Frequency/Density')

​

EDOS_mean = df['EDOS_Avg'].mean()

EDOS_std = df['EDOS_Avg'].std()

​

# Optional: Add vertical lines for mean and +/- 1 standard deviation

plt.axvline(EDOS_mean, color='r', linestyle='--', label=f'Mean: {EDOS_mean:.2f}')

plt.axvline(EDOS_mean - EDOS_std, color='g', linestyle=':', label=f'-1 Std Dev: {EDOS_mean - EDOS_std:.2f}')

plt.axvline(EDOS_mean + 2*EDOS_std, color='g', linestyle=':', label=f'+2 Std Dev: {EDOS_mean + 2*EDOS_std:.2f}')

plt.legend()

#  Making waterfall charts

!pip install waterfallcharts

import waterfall_chart as wc

import matplotlib.pyplot as plt

from waterfall_chart import plot as waterfall

from waterfallcharts import quick_charts as qc

data = {

    'Hospital': df['Hospital'],

    'ALOS_Avg': df['ALOS_Avg']

}

dfw = pd.DataFrame(data)

dfw.info()

import pandas as pd

import plotly.graph_objects as go

import plotly.express as px

​

# Sample data

​

data = {

    'Hospital': df['Hospital'],

    'ALOS_Avg': df['ALOS_Avg']

}

dfw = pd.DataFrame(data)

​

# Add the "Total" rows in their appropriate place within the dataframe

# For a simple cumulative total, you can add a final 'Total' category

# If specific intermediate totals are needed, a "measure" list is used

​

fig = go.Figure(go.Waterfall(

    name="ALOS",

    orientation="v",

    measure=["relative", "relative"], # All are changes relative to the previous

    x=dfw['Hospital'],

    y=dfw['ALOS_Avg'],

    connector={"line": {"color": "rgb(10, 20, 30)"}},

    increasing = {"marker":{"color":"Teal"}},

))

​

fig.update_layout(title="Siloam Hospitals: Avg ALOS for 2024",

                  xaxis_title="Hospital",

                  yaxis_title = "Average Length of Stay",

                  showlegend=True)

fig.show()

# Save the figure as an HTML file

fig.write_html("Siloam Hospitals: Avg ALOS for 2024.html")