實證的模型有時候會因為只採用某種研究方法，某種選取的基準，而獲得我們想要得到的結論，所以在此情況下此種實證結論往往是依…而定所產生的實證結果，不免可能會被懷疑，因為往往會令人感覺因實證結果太好而感覺不太真實，所以本實證文章乃是基於此情況下而來構思，亦為實證結果的強化性(Robustness)有其重要性，此亦本篇文章主要所探討的課題。本文的實證所採用的例子乃是貨幣波動性對通貨膨脹之影響為何，雖然這一方面的課題被不少學者所探討過，然而實證的強化性乃是本研究最主要的考量事項，與過去實證文章相比較，本文實證的所採行之構面，有變數不同的選取、不同模型之採行、各種落後期方法之採用、針對落後期對稱、不對稱亦有所考量、實際與模擬資料皆有採用，若是因而可以到相當一致性的結果，應該會較過去這方面的實證來的更具強化性。而實證結果顯示：即使不同變數挑選不同的落後期數，對稱模型與非對稱模型的實證結果相似，強化實證結果。 In empirical study, we might get the conclusion we want by applying specific methodologies or adapting special chosen criteria. Sometimes, the empirical results might be too good to be true, and be questioned since such empirical results are obtained under special conditions. Thus the robustness of empirical study seems crucial for empirical researches. In this paper, we investigate the effect of money uncertainty to inflation. Although this topic has been discussed by many scholars before, our paper has more robustness concerns than previous researches, and several concerns are included in this paper such as different substitute variables chosen, different models applied, different lag-length chosen criteria employed, symmetric and asymmetric models applied, real or simulated data used. Thus, if we still could get the quiet coincident results by the above concerns, then we could provide more robustness evidence than previous researches.
