| LDR | | 03173nmm uu200469 4500 |
| 001 | | 000000334071 |
| 005 | | 20240805175058 |
| 008 | | 181129s2018 |||||||||||||||||c||eng d |
| 020 | |
▼a 9780438373075 |
| 035 | |
▼a (MiAaPQ)AAI10827981 |
| 035 | |
▼a (MiAaPQ)uchicago:14415 |
| 040 | |
▼a MiAaPQ
▼c MiAaPQ
▼d 248032 |
| 082 | 0 |
▼a 310 |
| 100 | 1 |
▼a Chen, Likai. |
| 245 | 10 |
▼a Statistical Learning and High-Dimensional Inference for Time Dependent Data. |
| 260 | |
▼a [S.l.] :
▼b The University of Chicago.,
▼c 2018 |
| 260 | 1 |
▼a Ann Arbor :
▼b ProQuest Dissertations & Theses,
▼c 2018 |
| 300 | |
▼a 139 p. |
| 500 | |
▼a Source: Dissertation Abstracts International, Volume: 80-02(E), Section: B. |
| 500 | |
▼a Adviser: Wei Biao Wu. |
| 502 | 1 |
▼a Thesis (Ph.D.)--The University of Chicago, 2018. |
| 520 | |
▼a This thesis considers statistical learning, testing and inference for time dependent data. |
| 520 | |
▼a In the classical statistical learning theory, researchers primarily deal with independent data and there is a huge literature. In comparison, the case with time dependent data has been much less investigated. Concentration inequalities for supre |
| 520 | |
▼a For time series data the above problem has been much less studied and it becomes considerably more challenging since, in the presence of dependence, techniques and methods for independent settings cannot be directly applied. A popular way is to |
| 520 | |
▼a In the paper. |
| 520 | |
▼a Concentration Inequalities for Empirical Processes of Linear Time Series, co-authored with Wei Biao Wu, accepted by the Journal of Machine learning research, we gave an upper bound of T(z) without imposing strong mixing conditions, which is ver |
| 520 | |
▼a Besides the dependence, the rise of high-dimensional data brings new challenges to statistical inference. Statistical inference for the trends of high dimensional time series is essential in many areas. Consider the model with the observation (n |
| 520 | |
▼a In the literature, people make one or both of the following assumptions to perform inference on trends: (i) the dimension p is low, (ii) the processes are temporally or cross-sectionally independent. However, it is not uncommon that one needs t |
| 520 | |
▼a Testing for Trends in High-dimensional Time Series, co-authored with Wei Biao Wu, to appear on the Journal of the American Statistical Association was initially motivated by a temperature data gathered from various locations across America, and |
| 520 | |
▼a In our theory we relaxed both of above two restrictions by allowing a large p and temporal and cross-sectional dependencies. Based on a modified L2-distance between parametric and nonparametric trend estimates, we propose a de-diagonalized quadr |
| 590 | |
▼a School code: 0330. |
| 650 | 4 |
▼a Statistics. |
| 690 | |
▼a 0463 |
| 710 | 20 |
▼a The University of Chicago.
▼b Statistics. |
| 773 | 0 |
▼t Dissertation Abstracts International
▼g 80-02B(E). |
| 773 | |
▼t Dissertation Abstract International |
| 790 | |
▼a 0330 |
| 791 | |
▼a Ph.D. |
| 792 | |
▼a 2018 |
| 793 | |
▼a English |
| 856 | 40 |
▼u http://www.riss.kr/pdu/ddodLink.do?id=T14999095
▼n KERIS |
| 980 | |
▼a 201812
▼f 2019 |
| 990 | |
▼a 관리자 |