I build a small-scale dynamic stochastic general equilibrium (DSGE) model for Indonesia and assess which structural features of the model are consistent with actual data. I show that the best-fitting model does not include habit formation or backward-looking price indexation, which are widely assumed features in standard DSGE models. Money-holding friction, however, is found to be an important feature. When these features are assumed to be present, I find that the degrees of habit formation and price indexation decline after full-fledged adoption of the inflation targeting framework in Indonesia. This implies that agents of the economy believe that Bank Indonesia’s monetary policy is credible in achieving its inflation target. When using a large-scale DSGE model for policy analysis, central banks must evaluate which frictions should be included.

我为印度尼西亚建立了小型动态随机一般均衡（DSGE）模型，并评估了该模型的哪些结构特征与实际数据一致。我表明，最适合的模型不包括习惯形成或后向价格指数，这是标准DSGE模型中被广泛认为的特征。然而，发现持有货币的摩擦是一个重要特征。当假定这些特征存在时，我发现在印度尼西亚全面采用通货膨胀目标框架后，习惯形成和价格指数化的程度下降。这意味着经济主体认为印尼央行的货币政策在实现其通胀目标方面是可信的。当使用大规模DSGE模型进行政策分析时，中央银行必须评估应包括哪些摩擦。