公共衛(wèi)生科研項目：基于時間序列的數(shù)據(jù)挖掘和統(tǒng)計預(yù)測在衛(wèi)生疫情信息管理中的分析與研究

導(dǎo)師學(xué)校介紹

麻省理工學(xué)院（MIT）創(chuàng)立于1861年，是世界著名私立研究型大學(xué)，在2020年U.S.News世界大學(xué)排名中綜排位列第二。學(xué)校孕育了90位諾貝爾獎得主、59位美國國家科學(xué)獎?wù)芦@得者，以及75位麥克阿瑟獎獲得者。

導(dǎo)師詳細(xì)介紹

導(dǎo)師昵稱

Peter

導(dǎo)師級別

終身教職

導(dǎo)師學(xué)校

麻省理工學(xué)院(MIT)

Peter導(dǎo)師以優(yōu)異成績獲得哈佛大學(xué)(Harvard University)應(yīng)用數(shù)學(xué)學(xué)士學(xué)位，并當(dāng)選為Phi Beta Kappa Alpha Chapter的成員。后續(xù)他攻讀統(tǒng)計學(xué)碩士學(xué)位，并獲得了倫敦大學(xué)帝國理工學(xué)院(University of London)的碩士學(xué)位和文憑，以及加州大學(xué)伯克利分校(University of California Berkeley)的博士學(xué)位。在哈佛大學(xué)擔(dān)任統(tǒng)計學(xué)教授期間，他獲得了美國國家科學(xué)基金會的博士后數(shù)學(xué)科學(xué)研究獎學(xué)金。隨后，他成為麻省理工學(xué)院斯隆管理學(xué)院(MIT Sloan School of Management)的教授，并晉升為管理科學(xué)終身教授。從1990年到1998年，他還擔(dān)任麻省理工學(xué)院斯隆管理學(xué)院(MIT Sloan School of Management)的首席研究科學(xué)家，在經(jīng)濟和管理科學(xué)計算研究中心(CCREMS)和國際金融服務(wù)研究中心(IFSRC)進(jìn)行研究。他是風(fēng)險管理項目組的積極成員，并開發(fā)了納入行業(yè)標(biāo)準(zhǔn)RiskMetrics方法論的分析方法。2013年，他加入MIT數(shù)學(xué)系，擔(dān)任金融數(shù)學(xué)和統(tǒng)計講師。2014年在北京交通大學(xué)全球暑期學(xué)校任教期間，被聘為計算機與信息技術(shù)學(xué)院特聘教授。

自1992年以來，他一直通過他的公司Kempthorne analytics,Inc.為各種機構(gòu)提供金融和統(tǒng)計分析咨詢服務(wù)。過去的客戶包括花旗銀行(Citibank)、Colonial/Liberty Funds、美國運通(American Express)、巴黎國家銀行(Banque Nationale de Paris)、佳能(Canon)、富達(dá)管理與研究(Fidelity Management and Research)、Mathsoft/Corporation、默克(Merck)、RXR、山德士(Sandoz)和普林斯頓品牌計量經(jīng)濟學(xué)(Princeton Brand Econometrics)。項目活動包括:股票市場的資產(chǎn)選擇建模、風(fēng)險管理的統(tǒng)計分析、風(fēng)險管理軟件的集成設(shè)計和實現(xiàn)、衍生品定價的金融分析、災(zāi)難性風(fēng)險分析——風(fēng)險暴露建模和保險定價方案、用于做市的股票市場交易數(shù)據(jù)微觀結(jié)構(gòu)建模以及交易系統(tǒng)的設(shè)計、開發(fā)、實現(xiàn)。

自1995年以來，他一直擔(dān)任投資經(jīng)理，利用先進(jìn)的統(tǒng)計分析來管理各種投資項目。從2010年到2012年，他在IKOS,CIF Ltd擔(dān)任投資組合經(jīng)理和高級研究員，IKOS,CIF Ltd是一家完全系統(tǒng)化的量化對沖基金，管理著21億美元(美元)的全球股票、期貨和貨幣投資組合。作為投資組合經(jīng)理，管理和增強股票投資組合的實時構(gòu)建過程，包括alpha模型評估和開發(fā)、執(zhí)行分析和投資組合優(yōu)化;他還進(jìn)行了期貨和貨幣投資組合的風(fēng)險建模和管理。作為高級研究員，他擔(dān)任研究指導(dǎo)委員會主席，管理和指導(dǎo)研究人員，并協(xié)調(diào)IKOS/牛津大學(xué)博士實習(xí)生計劃。他于1995年聯(lián)合創(chuàng)立了Chronos Asset Management，并于1996年聯(lián)合創(chuàng)立了Summa Capital Management。作為這兩家投資管理公司的負(fù)責(zé)人，他運用自己專有的分析方法開發(fā)統(tǒng)計交易模型和交易系統(tǒng)，并監(jiān)督交易操作。

Kempthorne Analytics目前在馬薩諸塞州注冊為投資顧問，為零售客戶管理系統(tǒng)定量投資項目。Peter導(dǎo)師持有Series 3和Series 65許可證，并在the National Futures Association是注冊商品交易顧問。

他活躍于John Bertram House Inc.(1998-2010)和Lynn Home for Young Women,Inc.(2005-2010)的董事會。他曾擔(dān)任兩家非營利公司的財務(wù)主管，并擔(dān)任監(jiān)督信托資產(chǎn)管理的財務(wù)委員會主席。

Peter received his A.B.magna cum laude degree in applied mathematics from Harvard University and was elected to the Alpha Chapter of Phi Beta Kappa.He pursued graduate studies in statistics receiving the M.Sc.degree and the Diploma of Imperial College award from the University of London,and a Ph.D.from the University of California Berkeley.While an Assistant Professor of Statistics at Harvard,he was awarded a Postdoctoral Mathematical Sciences Research Fellowship by the National Science Foundation.He then joined the faculty of MIT at the Sloan School of Management as a visiting Assistant Professor and was promoted to Associate Professor of Management Science.From 1990 to 1998,he also served as a Principal Research Scientist at the MIT Sloan School of Management conducting research at the Center for Computational Research in Economics and Management Science(CCREMS)and at the International Financial Services Research Center(IFSRC).He was an active member of the Risk Management Working Group and developed analytics incorporated in the industry-standard RiskMetrics methodology.In 2013 Peter joined the MIT mathematics department as a lecturer in financial mathematics and statistics.His course""Topics in Mathematics with Applications to Finance""is published and available on the MIT Open Courseware website.In 2014,while teaching in the Global Summer School of Beijing Jiaotong University,he was appointed Distinguished Visiting Professor in the School of Computer and Information Technology.

Peter has been providing consulting services in financial and statistical analytics to a wide range of institutions through his company since 1992.Past clients include Citibank,Colonial/Liberty Funds,American Express,Banque Nationale de Paris,Canon,Fidelity Management and Research,Mathsoft/Insightful Corporation,Merck,RXR,Sandoz,and Princeton Brand Econometrics.Project activities include:asset selection modeling for equity markets,statistical analytics for risk management,integrated design and implementation of risk management software,financial analytics for derivatives pricing,catastrophic risk analytics-exposure modeling and pricing insurance programs,stock market microstructure modeling of transaction data for market making;and trading system design,development,and implementation.

Since 1995,Peter has been an investment manager,exploiting advanced statistical analytics to manage a variety of investment programs.From 2010-2012 he was portfolio manager and senior researcher at IKOS,CIF Ltd,a fully systematic,quantitative hedge fund managing$2.1B(USD)in global portfolios of equities,futures,and currencies.As portfolio manager he managed and enhanced real-time construction processes of equities portfolios,including alpha model evaluation and development,executions analysis and portfolio optimization;and he conducted risk modeling and management of futures and currency portfolios.As senior researcher he chaired the Research Steering Committee,managed and mentored researchers,and coordinated the IKOS/Oxford Univ.PhD intern program.He co-founded Chronos Asset Management in 1995 and Summa Capital Management in 1996.As a principal of both investment management companies,he applied his proprietary analytic methods to develop statistical trading models and trading systems and supervised trading operations.

Kempthorne Analytics is currently registered as an investment adviser in Massachusetts and manages systematic quantitative investment programs for retail clients.Peter holds the Series 3 and Series 65 licenses and is registered with the National Futures Association as a Commodity Trading Adviser.

Peter was active on the boards of John Bertram House Inc.(1998-2010)and the Lynn Home for Young Women,Inc.(2005-2010).He served as Treasurer for both non-profit corporations and chaired the finance committees that oversaw the management of trust assets.

適合人群

方向：金融商科

專業(yè)：經(jīng)濟學(xué)

適合專業(yè)：經(jīng)濟學(xué)，數(shù)據(jù)分析，抗擊冠狀病毒，數(shù)學(xué)，統(tǒng)計學(xué)，發(fā)展經(jīng)濟學(xué)，公共衛(wèi)生學(xué)，生物醫(yī)學(xué)統(tǒng)計，公共衛(wèi)生

項目價格：33800/19800

項目周期：7周在線小組科研+5周論文輔導(dǎo)

是否建議高中生學(xué)習(xí)：是

是否建議大學(xué)生學(xué)習(xí)：是

語言：英文

難度：中級/高級難度

建議具備的基礎(chǔ)：公共衛(wèi)生、衛(wèi)生信息統(tǒng)計、數(shù)學(xué)、商業(yè)管理、生物統(tǒng)計或數(shù)據(jù)預(yù)測等專業(yè)或希望修讀相關(guān)專業(yè)的學(xué)生；學(xué)生需具備隨機變量、概率論等相關(guān)知識并熟練掌握R語言

科研項目產(chǎn)出

7周在線小組科研學(xué)習(xí)+5周論文輔導(dǎo)學(xué)習(xí)共125課時

學(xué)術(shù)報告

優(yōu)秀學(xué)員獲主導(dǎo)師Reference Letter

EI/CPCI/Scopus/ProQuest/Crossref/EBSCO或同等級別索引國際會議全文投遞與發(fā)表指導(dǎo)（可用于申請）

結(jié)業(yè)證書

成績單

項目介紹

Introduction to fundamental methods and models of time series analysis with applications in economics,finance,and public health.Important models of trend and seasonality are developed and applied,using multi-stage exponential smoothing.Box-Jenkins models for stationary time series(auto-regressions,moving averages)are covered including methods for estimation,order selection,and forecasting.Real-world time series data are collected from the internet and analyzed with the methods covered in the program.

本項目將向?qū)W生介紹時間序列分析的基本方法和模型，以及在經(jīng)濟學(xué)、金融學(xué)和公共衛(wèi)生領(lǐng)域的應(yīng)用。利用多階段指數(shù)平滑，重要的趨勢和季節(jié)性模型得到了更好的發(fā)展和應(yīng)用。項目中介紹了用于固定時間序列（自回歸、移動平均線）的Box-Jenkins模型，包括估計、順序選擇和預(yù)測方法。學(xué)生們將從互聯(lián)網(wǎng)上收集現(xiàn)實世界的時間序列數(shù)據(jù)，并使用項目中涵蓋的方法進(jìn)行分析。

項目背景

時間序列是指將某種現(xiàn)象某一個統(tǒng)計指標(biāo)在不同時間上的各個數(shù)值，按時間先后順序排列而形成的序列。時間序列法是一種定量預(yù)測方法，亦稱簡單外延方法，在統(tǒng)計學(xué)中作為一種常用的預(yù)測手段被廣泛應(yīng)用。時間序列分析在第二次世界大戰(zhàn)前應(yīng)用于經(jīng)濟預(yù)測。二次大戰(zhàn)中和戰(zhàn)后，在軍事科學(xué)、空間科學(xué)、氣象預(yù)報和工業(yè)自動化等部門的應(yīng)用更加廣泛。時間序列分析(Time series analysis)是一種動態(tài)數(shù)據(jù)處理的統(tǒng)計方法。該方法基于隨機過程理論和數(shù)理統(tǒng)計學(xué)方法，研究隨機數(shù)據(jù)序列所遵從的統(tǒng)計規(guī)律，以用于解決實際問題。時間序列構(gòu)成要素是：現(xiàn)象所屬的時間，反映現(xiàn)象發(fā)展水平的指標(biāo)數(shù)值。

項目大綱介紹

時間序列分析導(dǎo)論 Introduction to Time Series Analysis

時間序列模型;金融時間序列 Simple Time Series Models;financial time series

預(yù)估噪聲序列的時間序列相關(guān)性檢驗固定的流程 Testing estimated noise sequences for time series dependence;stationary processes

回歸(AR)、移動平均(MA)和ARMA模型；模型選擇和預(yù)測

Auto-regression(AR),moving average(MA),and ARMA models；model selection and forecasting

學(xué)術(shù)研討1  Final Project Phase I

學(xué)術(shù)研討2  Final Project Phase II

項目回顧和成果展示 Program Review and Presentation

論文輔導(dǎo) Project Deliverables Tutoring

>>沒有教授推薦信，缺少科研經(jīng)驗？點擊這里提升背景<<