请用matlab写一个寻找arima最优参数的代码

下面是一个使用MATLAB寻找ARIMA最优参数的示例代码：

% 导入时间序列数据
data = readtable('data.csv');
y = data.Value;

% 寻找最优参数
best_aic = Inf;
best_order = [0 0 0];
best_seasonal_order = [0 0 0 0];
for p = 0:3
    for d = 0:1
        for q = 0:3
            for s = 0:1
                if s == 0
                    seasonal_order = [0 0 0 0];
                else
                    seasonal_order = [1 0 0 12];
                end
                try
                    mdl = arima(p,d,q);
                    est_mdl = estimate(mdl, y, 'Seasonality', seasonal_order);
                    aic = est_mdl.ModelCriterion.AIC;
                    if aic < best_aic
                        best_aic = aic;
                        best_order = [p d q];
                        best_seasonal_order = seasonal_order;
                    end
                catch
                    continue
                end
            end
        end
    end
end

% 输出最优参数
fprintf('Best ARIMA Model: ARIMA(%d,%d,%d)\n', best_order);
if best_seasonal_order(1) == 0
    fprintf('Seasonal Component: None\n');
else
    fprintf('Seasonal Component: ARIMA(%d,%d,%d)(%d)\n', best_seasonal_order);
end
fprintf('AIC: %f\n', best_aic);

% 拟合最优ARIMA模型并进行预测
mdl = arima(best_order(1), best_order(2), best_order(3));
est_mdl = estimate(mdl, y, 'Seasonality', best_seasonal_order);
[y_pred, y_pred_ci] = forecast(est_mdl, 12);

% 绘制预测结果图像
figure;
plot(y);
hold on;
plot(y_pred, 'r');
plot(y_pred_ci, 'r--');
xlabel('Time');
ylabel('Value');
legend('Observed', 'Forecast', '95% Confidence Interval');

请注意，此代码仅用于示例目的，实际数据可能需要使用不同的ARIMA模型。同时，代码中的参数范围也可能需要根据实际情况进行调整。