赛题:
问题2:
针对康复医学科二病房,以日期、患者人数、药品总收入、当日病房收入为基础输入数据,构建2个不同类型的时间序列预测模型,分别预测对应的门诊收入数据,目标为表中这个科室空白的数据。
解题思路:
第2题的解题思路跟第1题完全一样,可以回去翻看我前面发的思路
代码:
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
from statsmodels.tsa.arima.model import ARIMA
from sklearn.ensemble import RandomForestRegressor
from sklearn.model_selection import train_test_split
from sklearn.metrics import mean_squared_error
from sklearn.preprocessing import StandardScaler
# 1. 加载数据
data = pd.read_excel('数据.xlsx')
# 2. 数据预处理
# 选择康复医学科二病房的数据
ward_data = data[data['当日病房收入对应科室'] == '康复医学科二病房']
# 填充缺失值
ward_data.fillna(method='ffill', inplace=True)
# 创建有用的特征
ward_data['日期'] = pd.to_datetime(ward_data['日期'])
ward_data['Month'] = ward_data['日期'].dt.month
ward_data['DayOfWeek'] = ward_data['日期'].dt.dayofweek
# 准备训练和测试数据
X = ward_data[['门诊患者人次数', '药品总收入', '当日病房收入', 'Month', 'DayOfWeek']]
y = ward_data['门诊收入OBS_T01_MZSR68']
# 3. 探索性数据分析
plt.figure(figsize=(14, 7))
plt.plot(ward_data['日期'], ward_data['门诊收入OBS_T01_MZSR68'])
plt.title('康复医学科二病房门诊收入随时间的变化')
plt.xlabel('日期')
plt.ylabel('门诊收入')
plt.show()
# 4. 模型构建
# 4.1 ARIMA 模型
# 选择 p, d, q 参数
p = 1
d = 1
q = 1
# 4.2 随机森林模型
rf = RandomForestRegressor(n_estimators=100, random_state=42)
# 5. 模型训练与评估
# 划分训练集和测试集
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)
# ARIMA 模型训练
# 需要将时间序列数据转换为监督学习格式
def series_to_supervised(data, n_in=1):
df = pd.DataFrame(data)
cols = list()
for i in range(n_in, 0, -1):
cols.append(df.shift(i))
cols.append(df)
agg = pd.concat(cols, axis=1)
agg.dropna(inplace=True)
return agg.values
values = ward_data['门诊收入OBS_T01_MZSR68'].values
supervised_data = series_to_supervised(values, 1)
train = supervised_data[:int(len(supervised_data)*0.8)]
test = supervised_data[int(len(supervised_data)*0.8):]
history = [x for x in train]
predictions = list()
for t in range(len(test)):
model = ARIMA(history, order=(p,d,q))
model_fit = model.fit()
output = model_fit.forecast()
yhat = output[0]
predictions.append(yhat)
obs = test[t]
history.append(obs)
print('predicted=%f, expected=%f' % (yhat, obs))
# 计算 MSE
mse = mean_squared_error(test[:, 0], predictions)
print('ARIMA Test MSE: %.3f' % mse)
# 随机森林模型训练
scaler = StandardScaler()
X_train_scaled = scaler.fit_transform(X_train)
X_test_scaled = scaler.transform(X_test)
rf.fit(X_train_scaled, y_train)
# 6. 评估模型性能
y_pred_arima = predictions
y_pred_rf = rf.predict(X_test_scaled)
# 计算 MSE
mse_arima = mean_squared_error(y_test, y_pred_arima)
mse_rf = mean_squared_error(y_test, y_pred_rf)
print('ARIMA Test MSE: %.3f' % mse_arima)
print('Random Forest Test MSE: %.3f' % mse_rf)