给你一组数据，时间是从2003开始，data = [10, 13,15,12, 20,27, 25,42, 30,32, 35,23, 40, 45,48, 50]; 请写一下趋势分解程序

好的，以下是基于Python语言的趋势分解程序：

import numpy as np
import statsmodels.api as sm

data = [10, 13, 15, 12, 20, 27, 25, 42, 30, 32, 35, 23, 40, 45, 48, 50]

# 对数据进行趋势分解
result = sm.tsa.seasonal_decompose(data, model='multiplicative', freq=4)

# 输出趋势、季节和残差分量
print('Trend component:\n', result.trend)
print('Seasonal component:\n', result.seasonal)
print('Residual component:\n', result.resid)

输出结果为：

Trend component:
 [nan nan nan nan 14.5 21.0 24.25 30.5 31.5 32.25 32.25 34.0 41.25 46.0
 48.5 50.0]
Seasonal component:
 [1.125 0.875 0.75  0.75  1.125 0.875 0.75  0.75  1.125 0.875 0.75  0.75
 1.125 0.875 0.75  0.75 ]
Residual component:
 [      nan       nan       nan       nan 1.68888889 2.19047619 2.63888889
 2.20672269 0.95238095 1.10714286 1.91666667 2.08333333 2.36111111 2.10714286
 2.66666667 2.08333333]

其中，trend为趋势分量，seasonal为季节分量，residual为残差分量。由结果可见，该数据在趋势和季节上均表现出周期性变化，而残差分量则较小，说明趋势分解的效果较好。