在這技術發達的時代，穿戴式裝置成為是一項新穎的智能電子設備，配戴於身上能即時且準確地收集人類身體的活動訊息。此裝置所收集到的資料型態多可視為函數型資料。而在統計上，一個常見的方法-----函數型資料分析(Functional Data Analysis, FDA)便常用於處理此類型的資料。在本文中，將利用三步驟的方式挑選出影響運動強度數據的重要變數。首先，將運動強度數據轉換成運動強度剖面以解決資料具有缺失值的問題；接著，利用隨機投影的方法來降低資料維度，以增加統計分析上的便利性；最終，結合向前選取法(Forward Selection)與變異數分析(ANOVA)來挑選出影響人們平時運動強度的主要因素。我們透過三個模擬實驗來驗證此論文所採用方法之有效性與實用性，並應用至美國國家健康與營養調查2005-2006年的穿戴式裝置之資料集上。本研究發現，影響美國老年人之運動強度的重要變數分別為性別和年齡。

In this technologically advanced age, wearable devices have become a novel intelligent electronic equipment that can be worn on the body to collect information about human activities immediately and accurately.
The data collected by wearable devices can be regarded as functional data.
In this thesis, we use a three-step method to select important variables that affect activity intensity data.
First, we convert the activity intensity data into activity profile to align the data and to deal with missing values.
Second, we use the random projection method to reduce the dimension which increases the convenience of statistical analysis.
Finally, we select important variables that affect people's activity intensity by forward selection and analysis of variance.
Three simulation experiments are provided to show the validity and power of the proposed method.
We apply our method on the wearable device data set of the National Health and Nutrition Survey of the United States 2005-2006. We find that gender and age are both significantly affect elder citizen's activity intensity.

1. Benjamini, Y., & Yekutieli, D. (2001). The control of the false discovery rate in multiple testing under dependency. The Annals of Statistics, 29(4), 11651188.
2. CuestaAlbertos,J. A., & FebreroBande,M. (2010). A simple multiway ANOVA for functional data. Test, 19(3), 537-557.
3. G´orecki, T. & Smaga, L. (2017). Multivariate analysis of variance for functional data. Journal of Applied Statistics, 44(12), 2172-2189.
4. Healy, G. N., Dunstan, D. W., Salmon, J., Cerin, E., Shaw, J. E., Zimmet, P. Z., & Owen, N.
(2007). Objectively measured light-intensity physical activity is independently associated with 2-h plasma glucose. Diabetes Care, 30(6), 1384-1389.
5. Masse, L.C., Fuemmeler, B.F., Anderson, C.B., Matthews, C.E., Trost, S.G., Catellier, D.J.,
and Treuth, M.(2005). Accelerometer data reduction: a comparison of four reduction algorithms on select outcome variables. Medicine and Science in Sports and Exercise, 37, S544-554.
6. Matthews, C. E., Chen, K. Y., Freedson, P. S., Buchowski, M. S., Beech, B. M., Pate, R. R.,
& Troiano, R. P. (2008). Amount of time spent in sedentary behaviors in the United States, 2003–2004. American Journal of Epidemiology, 167(7), 875-881.
7. Rowlands, A. V., Edwardson, C. L., Davies, M. J., Khunti, K., Harrington, D., & Yates, T. (2018). Beyond cut points: Accelerometer metrics that capture the physical activity profile. Medicine & Science in Sports & Exercise, 50(6), 1323-1332.
8. Treuth, M. S., Schmitz, K., Catellier, D. J., McMurray, R. G., Murray, D. M., Almeida, M. J.,& Pate, R. (2004). Defining accelerometer thresholds for activity intensities in adolescent girls. Medicine and Science in Sports and Exercise, 36(7), 1259.
9. Troiano, R. P. (2006). Translating accelerometer counts into energy expenditure: advancing
the quest. Journal of Applied Physiology, 100, 1107-1108.
10. 財團法人資訊工業策進會(2017). 智慧穿戴，創新生活，解析智慧穿戴式裝置市場趨勢與創新產品應用. Retrieved 2020/02/03, from https://www.iii.org.tw/Focus/FocusDtl.aspx?f_type=2&f_sqno=Zdw7bw%2B50oGAn2GA6qArNg__&fm_sqno=13