多部门季度数据预测：Plotly多Y轴折线图报错修复

需要对Fin、Legal、Leadership、Overall等部门的季度数据进行时间序列预测，直到2023年第四季度。尝试用Plotly绘制多变量折线图时触发ValueError，错误信息如下：

ValueError: All arguments should have the same length. The length of argument y is 4, whereas the length of previously-processed arguments ['Time Period'] is 6

示例数据 #

报错代码 #

import pandas as pd
from datetime import date, timedelta
import datetime
import matplotlib.pyplot as plt
plt.style.use('fivethirtyeight')
from statsmodels.tsa.seasonal import seasonal_decompose
from statsmodels.graphics.tsaplots import plot_pacf
from statsmodels.tsa.arima_model import ARIMA
import statsmodels.api as sm
import warnings
import plotly.graph_objects as go
from plotly.subplots import make_subplots

inputfilepath = 'C:/Documents/Forecast/Input/Forecast Data csv.csv'
df = pd.read_csv(inputfilepath)
print(df)

import plotly.express as px
figure = px.line(df, x="Time Period", 
                 y=("Fin","Legal","Leadership","Overall"),
                 title='Quarterly scores')

figure.show()

错误原因 #

报错核心是DataFrame列名不匹配：示例数据第一列的实际列名为Time Period HR（疑似输入笔误），但代码中指定的x轴列名为Time Period，导致Plotly无法找到对应列，进而误判数据长度不匹配。

修复后的绘图代码 #

先修正列名，再绘制多变量折线图：

import pandas as pd
import plotly.express as px

# 读取数据
inputfilepath = 'C:/Documents/Forecast/Input/Forecast Data csv.csv'
df = pd.read_csv(inputfilepath)

# 修正列名（匹配代码中使用的x轴字段）
df.rename(columns={"Time Period HR": "Time Period"}, inplace=True)

# 绘制多Y轴折线图
figure = px.line(df, x="Time Period", 
                 y=["Fin", "Legal", "Leadership", "Overall"],
                 title='各部门季度得分趋势')
# 优化轴标签显示
figure.update_layout(xaxis_title="时间周期", yaxis_title="得分")
figure.show()

扩展：实现2023Q4前的时间序列预测 #

针对季度季节性数据，使用SARIMA模型完成预测，完整代码如下：

import pandas as pd
import warnings
import plotly.graph_objects as go
from statsmodels.tsa.statespace.sarimax import SARIMAX

warnings.filterwarnings("ignore")

# 1. 数据预处理
inputfilepath = 'C:/Documents/Forecast/Input/Forecast Data csv.csv'
df = pd.read_csv(inputfilepath)
df.rename(columns={"Time Period HR": "Time Period"}, inplace=True)

# 将季度字符串转为datetime格式，设置时间索引
df['Time Period'] = pd.to_datetime(df['Time Period'].str.replace('Q', '-') + '-1', format='%Y-%m-%d')
df.set_index('Time Period', inplace=True)
df = df.asfreq('Q')  # 明确季度频率

# 2. 定义单部门预测函数（适配季度季节性）
def forecast_dept(data_col, steps=5):
    # SARIMA(1,1,1)(1,1,1,4)：4代表季度周期
    model = SARIMAX(data_col, order=(1,1,1), seasonal_order=(1,1,1,4))
    result = model.fit()
    # 生成预测值与置信区间
    forecast = result.get_forecast(steps=steps)
    return forecast.predicted_mean, forecast.conf_int()

# 3. 预测各部门至2023Q4（现有数据到2022Q3，需预测5个季度）
departments = ["Fin", "Legal", "Leadership", "Overall"]
forecast_data = {}
for dept in departments:
    forecast_data[dept] = forecast_dept(df[dept], steps=5)

# 4. 绘制含预测值的多折线图
fig = go.Figure()

# 添加历史数据轨迹
for dept in departments:
    fig.add_trace(go.Scatter(x=df.index, y=df[dept], mode='lines+markers', name=f'{dept} 历史数据'))

# 添加预测数据及置信区间
for dept in departments:
    pred_vals, conf_int = forecast_data[dept]
    fig.add_trace(go.Scatter(x=pred_vals.index, y=pred_vals, mode='lines+markers', 
                             name=f'{dept} 预测数据', line=dict(dash='dash')))
    # 填充置信区间
    fig.add_trace(go.Scatter(
        x=pred_vals.index.append(pred_vals.index[::-1]),
        y=conf_int.iloc[:,0].append(conf_int.iloc[:,1][::-1]),
        fill='toself', fillcolor='rgba(0,0,0,0.1)',
        line=dict(color='rgba(255,255,255,0)'),
        name=f'{dept} 置信区间'
    ))

# 优化图表布局
fig.update_layout(title='各部门季度得分及预测（至2023Q4）',
                  xaxis_title='时间周期', yaxis_title='得分',
                  legend=dict(orientation='h', yanchor='bottom', y=1.02, xanchor='right', x=1))
fig.show()

内容的提问来源于stack exchange，提问作者Scythor