• 沒有找到結果。

3.4.2 Clark and West檢定

本研究採用 Clark and West(2007) 提出之檢定方法, 評估精簡模型 (parsi-monious model) 以及含有較多參數之結構模型 (large model)之預測表現。

在本研究中我們以隨機漫步模型代表精簡模型,因子結合其他基本面模型代

‧ 國

立 政 治 大 學

N a tio na

l C h engchi U ni ve rs it y

4 實證結果

在本節中, 我們將使用 Engel, Mark and West (2014) 之因子模型以及因子 結合其他前述所提及之匯率預測模型, 透過遞迴法進行樣本外預測, 並利用

Theil’s U 比例及 CW 檢定比較各模型對於匯率預測之準確性。 其中, 由於

考慮到歐元於199911日正式啟用以及其因調整所需之過渡期間,因此 將資料期間設定為20001月至20179月。 爾後,我們將探討全球因子及 亞洲區因子對於亞洲匯率之預測表現差異,以及各幣別對於不同模型之預測 表現。

4.1

全球因子對匯率之預測表現

2為全球因子 (我們採用 Kaiser 準則所選取之因子個數為4因子) 對匯率 相對於隨機漫步之預測表現, 其中, U 值之中位數隨著期數拉長, 數值越小, 顯示各國貨幣之長期預測表現優於短期。 表2中顯示, 購買力平價在極長期 (h=36)預測下表現為最好,其中位數為0.969且在17個樣本中,10個樣本 中位數小於1, 且有5個樣本顯著拒絕 CW 檢定之虛無假設。 此結果顯示全 球匯率在極長期 (h=36) 預測下, 購買力平價模型之預測表現較可能優於隨 機漫步模型,其中中位數小於1且顯著之國家為瑞典、 台灣、 印尼、 馬來西亞 與新加坡。 在中長期 (h=1224)的預測區間中,以貨幣學派模型及泰勒模型 表現較佳。 而以各幣別來看, 丹麥克朗以因子結合貨幣學派理論做為其預測 模型之表現為最佳,其在中長期(h=122436)之預測表現皆優於隨機漫步, 港幣則以因子結合泰勒法則做為其預測模型之表現為最佳,在短期(h=6)

極長期(h=36)的區間皆有優於隨機漫步之預測表現。

No.currency 6 12 24 36

Fˆi t si t Global/ median U 2.167 1.639 1.234 1.042

No.currency 6 12 24 36

Fˆi t si t Global/ median U 2.194 1.665 1.277 1.09

‧ 國

立 政 治 大 學

N a tio na

l C h engchi U ni ve rs it y

4:

亞洲因子預測表現

(r = 3)

model Sample

statistic 預測區間(h)

No.currency 6 12 24 36

Fˆi t si t Asian/ median U 2.251 1.723 1.326 1.165 N =10 U < 1 0(0) 0(0) 3(2) 3(1) Fˆi t si t+PPP Asian/ median U 3.509 2.919 2.048 1.971

N =10 U < 1 0(0) 1(1) 3(3) 4(4) Fˆi tsi t+Monetary Asian/ median U 3.9 2.586 1.545 1.39 N =9 U < 1 0(0) 0(0) 3(1) 2(0) Fˆi t si t+Taylor Asian/ median U 3.062 2.163 1.754 1.579

N =9 U < 1 0(0) 0(0) 0(0) 1(1) Fˆi t si t+UIP Asian/ median U 2.353 1.607 1.525 1.806

N =9 U < 1 0(0) 0(0) 1(1) 1(1) Note: 同表2說明。

54因子之預測表現,結果顯示,在中長期之預測表現皆有明顯優於3 因子之表現,且仍以購買力平價模型表現為佳;亞洲各國之貨幣在長期下,以 泰勒法則及未拋補利率平價作為預測之模型皆有顯著改善; 就各幣別而言, 台幣在購買力平價模型中仍保持優異之預測表現,韓元在4因子模型中顯著 優於3因子模型,且在中長期表現皆優於隨機漫步模型。

No.currency 6 12 24 36

Fˆi t si t Asian/ median U 2.312 1.701 1.142 1.103

Model Sample/No.currency statistic 6 12 24 36

Fˆi t si t Global/N = 10 2.48 1.729 1.386 1.103 0(0) 1(1) 3(2) 4(0)

Asian/N = 10 2.312 1.701 1.142 1.103

0(0) 1(0) 3(1) 5(2) Fˆi t si t+PPP Global/N = 10 2.258 1.605 1.161 0.843

1(0) 2(0) 3(3) 6(4)

Asian/N = 10 2.365 1.839 1.357 1.065

0(0) 1(0) 4(2) 5(3) Fˆi tsi t+Monetary Global/N = 9 3.706 2.306 1.608 1.376

medianU 0(0) 1(0) 1(1) 1(2) Asian/N = 9 U < 1 2.957 2.245 1.758 1.536

1(0) 1(1) 0(0) 1(2) Fˆi t si t+Taylor Global/N = 9 2.885 2.005 1.5 1.025

0(0) 1(1) 2(1) 4(2)

Asian/N = 9 2.827 2.009 1.412 1.117

0(0) 1(1) 1(1) 3(4) Fˆi t si t+UIP Global/N = 9 2.798 1.823 1.608 2.039

0(0) 0(0) 0(0) 3(2)

Asian/N = 9 2.499 1.736 1.659 1.202

0(0) 1(0) 2(1) 1(0)

Country Factor h=6 h=12 h=24 h=36

泰銖 global 3.6606 1.8467 0.8836 0.7061

asian 3.0158 2.1581 1.5737 1.2921

日圓 global 2.7179 1.9481 1.4711 0.9465

asian 2.6315 1.8555 1.1725 0.7968

台幣 global 1.9592 1.4856 1.1843 0.9197

asian 1.9916 1.5459 1.1116 0.9135

港幣 global 1.4158 1.24 0.9611 1.152

asian 1.4087 1.2539 1.0959 1.3124

韓元 global 1.45 0.8489 0.7146 0.7643

asian 1.064 0.9695 0.7567 0.6459

購買力平價模型(Purchasing power parity model)

Country Factor h=6 h=12 h=24 h=36

泰銖 global 2.1022 1.5043 1.1103 0.7473

asian 2.9902 2.1397 1.5268 1.1929

日圓 global 2.4178 1.7048 1.151 0.7514

asian 2.2038 1.5539 0.9926 0.6947

台幣 global 1.2739 0.9888 0.737 0.629

asian 1.2026 0.9335 0.6858 0.5685

港幣 global 1.4542 1.2944 1.1788 1.4187

asian 2.0258 1.8032 1.6165 1.9436

韓元 global 0.9575 0.8725 0.7626 0.77

asian 1.2074 1.1001 0.8619 0.9375

Country Factor h=6 h=12 h=24 h=36

泰銖 global 10.7207 2.5208 3.5168 3.1347

asian 2.0169 1.4662 3.6242 3.226

日圓 global 3.2156 2.5099 1.6076 1.1073

asian 3.7842 2.9255 1.5843 1.0993

台幣 global 2.6082 2.3064 1.6002 1.3763

asian 2.0954 2.1508 1.7579 1.5362

港幣 global 1.1767 0.9812 0.6328 0.7292

asian 0.9881 0.7608 1.0327 1.2211

韓元 global 5.4277 2.2423 2.206 2.3373

asian 6.4381 2.5934 3.9216 4.2491

泰勒法則模型(Taylor rule model)

Country Factor h=6 h=12 h=24 h=36

泰銖 global 3.2344 1.2125 1.8167 1.6065

asian 4.246 0.8598 1.8969 1.6847

日圓 global 2.8849 2.0072 1.5003 1.0255

asian 2.8266 2.1868 1.4121 0.9634

台幣 global 1.9314 1.4605 1.0573 0.7788

asian 1.7182 1.3885 1.1103 0.8312

港幣 global 1.0422 0.7161 0.8158 0.9693

asian 1.0971 1.158 0.8543 1.0109

韓元 global 1.6122 1.1936 0.8415 0.8973

asian 1.5238 1.1692 1.0471 1.1174

Country Factor h=6 h=12 h=24 h=36

泰銖 global 4.2636 2.7604 2.4162 3.1417

asian 3.3013 2.6545 2.4 2.7186

日圓 global 1.9864 1.6904 1.5313 1.07

asian 2.2165 1.7357 1.7096 1.2016

台幣 global 2.6188 1.8234 1.1458 0.9118

asian 1.7675 1.1726 0.9827 1.0464

港幣 global 1.4925 1.3231 1.0256 0.9744

asian 1.3807 1.5502 1.054 0.9451

韓元 global 1.4108 1.2728 1.62 2.0392

asian 1.2656 0.9072 0.7451 1.0408

Note:(1)global表示以全球匯率估計出之因子,並選擇因子個數為5,asian表示以亞洲匯率

Model Sample statistic 6 12 24 36

Fˆi t si t early 2.417 1.767 0.913 0.89 0(0) 3(1) 5(5) 5(3)

late 1.362 1.162 0.993 0.728

1(0) 4(3) 5(4) 6(3) Fˆi t si t+PPP early 3.157 2.073 1.268 1.054

0(0) 1(0) 4(4) 4(0)

late 1.774 1.463 1.014 0.72

0(0) 0(0) 4(3) 6(5) Fˆi t si t+Monetary early 3.247 2.292 1.648 1.353

medianU 0(0) 2(0) 3(3) 2(3) late U < 1 2.259 1.84 1.42 0.882

1(1) 3(3) 3(3) 5(4) Fˆi t si t+Taylor early 2.348 1.34 1.296 1.113

2(2) 4(4) 3(3) 4(3)

late 2.14 1.351 1.07 0.686

2(1) 4(2) 4(4) 7(4) Fˆi tsi t+UIP early 1.633 1.225 1.043 1.054

1(1) 4(3) 6(4) 6(1)

late 1.712 1.522 1.188 2.024

1(1) 1(1) 2(1) 1(1)

Model Sample statistic 6 12 24 36

Fˆi t si t early 2.643 1.898 0.914 1.032 0(0) 0(0) 5(5) 5(5)

late 1.309 1.131 1.034 1.48

2(0) 3(2) 4(3) 3(2) Fˆi t si t+PPP early 2.866 1.969 1.45 1.052

0(0) 1(1) 4(4) 5(1)

late 1.678 1.458 1.115 1.049

1(0) 1(1) 4(3) 4(3) Fˆi t si t+Monetary early 3.046 2.48 1.779 1.666

medianU 0(0) 1(1) 3(3) 2(2) late U < 1 2.621 1.594 1.385 1.604

2(0) 2(2) 3(3) 4(4) Fˆi t si t+Taylor early 2.71 1.229 1.252 1.249

3(2) 4(3) 3(3) 3(3)

late 1.968 1.32 1.02 1.075

1(1) 2(2) 4(3) 3(3) Fˆi tsi t+UIP early 1.651 1.537 1.071 1.1

2(0) 2(0) 4(3) 2(2)

late 1.823 1.493 1.299 1.916

1(1) 1(1) 2(2) 1(1) Note:同表2說明。

Berkowitz, J. and Giorgianni, L. (2001). “Long-horizon Exchange Rate Predictability?” Review of Economics and Statistics, 83, 81-91.

Corte, P. D., Sarno, L. and Tsiakas, I. (2011). “Spot and forward volatility in foreign exchange.” Journal of Financial Economics, 100, 496-513.

Engel, C., Mark, N. C. and West, K. D. (2015). “Factor model forecasts of exchange rates.” Econometric Reviews, 34, 32-55.

Frenkel, J. A. (1976). “A monetary approach to the exchange rate.” Scandi-navian J. Econ. 78, no.2, 200-224.

Groen, J. J. J. (2005). “Exchange rate predictability and monetary funda-mentals in a small multi-country panel.” Journal of Money, Credit and Banking, 37, 495-516.

Groen, J. J. J. (2006). “Fundamentals based exchange rate prediction revis-ited.” Manuscript, Bank of England.

Hodrick, R. J. and Prescott, E. C. (1981). “Post-war U.S. business cycles:

an empirical investigation.” Journal of Money, Credit and Banking, Vol. 29, No.1, 1-16.

Mark, N. C. (1995). “Exchange rate and fundamentals: evidence on long-horizon predictability.” American Economic Review, 85, 201-218.

Mark, N. C. and Sul, D. (2001). “Nominal exchange rates and monetary fundamentals: Evidence from a small post-Bretton Woods sample.”

‧ 國

立 政 治 大 學

N a tio na

l C h engchi U ni ve rs it y

Journal of International Economics, 53, 29-52.

Meese, R. A. and Rogoff, K. (1983). “Empirical exchange rate models of the seventies:do they fit out of sample.” Journal of International Economics, 14, 3-24.

Pincheira, P. and Gatty, A. (2014). “Forecasting chilean inflation with in-ternational factors.” Working Paper 723, Central Bank of Chille.

Rapach, D. E. and Wohar, M. E. (2004). “Testing the monetary model of exchange rate determination: A closer look at panels.” Journal of International Money and Finance, 23(6), 841-865.

Stock, J. H. and Watson, M. W. (2002). “Macroeconomic forecasting using diffusion indexes.” Journal of Business and Economic Statistics, 20, 147-162.

Stock, J. H. and Watson, M. W. (2006). “Forecasting with many predic-tors.” ch.6 in Handbook of Economic Forecasting, ed. by Elliott, G., Granger, C. and Timmerman A. Elsevier, 515-554.

West, K. D. and Wong, K. F. (2014). “A factor model for co-movements of commodity prices.” Journal of International Money and Finance, 42, 289-309.

相關文件