本文選題：函數(shù)型數(shù)據(jù) + 半?yún)?shù)模型　； 參考：《合肥工業(yè)大學(xué)》2017年碩士論文

【摘要】：隨著時代發(fā)展及科學(xué)技術(shù)進(jìn)步,研究人員在收集數(shù)據(jù)的途徑得到擴(kuò)展、能力得到提高,收集到的一些數(shù)據(jù)具有明顯的函數(shù)型的特點。例如:食品行業(yè)豬肉的光譜數(shù)據(jù)、電力行業(yè)日均負(fù)荷數(shù)據(jù)、多個地區(qū)的多個指標(biāo)的經(jīng)濟(jì)數(shù)據(jù)以及地區(qū)氣溫、降水量等數(shù)據(jù)。因此,對于函數(shù)型數(shù)據(jù)的研究已經(jīng)成為國內(nèi)外數(shù)理統(tǒng)計學(xué)界的研究的熱點之一。本學(xué)位論文主要由局部線性回歸下研究半函數(shù)型部分線性回歸模型的漸近性質(zhì)以及函數(shù)型數(shù)據(jù)在實際生活中的應(yīng)用組成,主要內(nèi)容如下:一、基于局部線性的半函數(shù)型部分線性回歸模型的估計在半函數(shù)型部分線性回歸模型下,通過局部線性回歸的方法對模型中函數(shù)型非參數(shù)部分進(jìn)行新的估計,得到參數(shù)及非參數(shù)部分的估計量,并探究參數(shù)部分估計量的漸近正態(tài)性、函數(shù)型非參數(shù)估計量的幾乎處處收斂速度。最后給出模擬分析,并與經(jīng)典的Nadarage-Watson核估計方法進(jìn)行比較,通過箱型圖表現(xiàn)出在局部線性方法要比核估計方法的效果好;對于參數(shù)部分我們通過直方圖檢驗估計量的漸近正態(tài)性。二、基于函數(shù)型非參數(shù)方法的氣溫數(shù)據(jù)分析預(yù)測在函數(shù)型非參數(shù)模型下,我們通過Nadarage-Watson核估計方法分析安徽省1955年1月至2010年12月月度平均氣溫數(shù)據(jù),建立函數(shù)型非參數(shù)回歸模型,并對2010年氣溫數(shù)據(jù)進(jìn)行實證分析研究。同時,從預(yù)測值的均方方差以及預(yù)測的氣溫曲線兩方面來對有限維非參數(shù)回歸模型及函數(shù)型非參數(shù)模型進(jìn)行比較,從而得出函數(shù)型非參數(shù)模型的優(yōu)越性。
[Abstract]:With the development of the times and the progress of science and technology, the way and the ability of the researchers to collect the data have been expanded, and some of the collected data have obvious functional characteristics.For example: the spectral data of pork in food industry, the daily load data of power industry, the economic data of multiple indexes in many regions, and the data of regional temperature, precipitation and so on.Therefore, the study of functional data has become one of the hotspots in the field of mathematical statistics at home and abroad.This dissertation is mainly composed of the asymptotic properties of semi-functional partial linear regression model under local linear regression and the application of functional data in real life. The main contents are as follows: 1.Based on the estimation of locally linear partial linear regression model in the semi-functional partial linear regression model, a new estimation of the functional nonparametric part of the model is carried out by means of the local linear regression method.The estimators of parametric and nonparametric parts are obtained, and the asymptotic normality of parametric partial estimators and almost everywhere convergence rate of functional nonparametric estimators are discussed.Finally, the simulation analysis is given, and compared with the classical Nadarage-Watson kernel estimation method, the box diagram shows that the local linear method is more effective than the kernel estimation method, and for the parameter part, we test the asymptotic normality of the estimator by histogram.Second, the temperature data analysis and prediction based on the functional nonparametric method. Under the functional nonparametric model, we analyze the monthly mean temperature data of Anhui Province from January 1955 to December 2010 by using the Nadarage-Watson kernel estimation method.A functional non-parametric regression model was established and the temperature data in 2010 were analyzed.At the same time, the finite dimensional nonparametric regression model and the functional nonparametric model are compared from the mean square variance of the predicted values and the predicted temperature curve, and the superiority of the functional nonparametric model is obtained.
【學(xué)位授予單位】：合肥工業(yè)大學(xué)
【學(xué)位級別】：碩士
【學(xué)位授予年份】：2017
【分類號】：O212.1

【參考文獻(xiàn)】

相關(guān)期刊論文 前8條

1 王定成;曹智麗;陳北京;倪郁佳;;日氣溫多元時間序列局部支持向量回歸預(yù)測[J];系統(tǒng)仿真學(xué)報;2016年03期

2 吳波;;基于遍歷函數(shù)型數(shù)據(jù)條件風(fēng)險率函數(shù)的核估計[J];大學(xué)數(shù)學(xué);2015年06期

3 許梁;孫濤;徐箭;孫元章;李子壽;林常青;;基于函數(shù)型非參數(shù)回歸模型的中長期日負(fù)荷曲線預(yù)測[J];電力自動化設(shè)備;2015年07期

4 陳百碩;李守偉;何建敏;曹杰;;天氣衍生品中時變均值回復(fù)的氣溫預(yù)測模型研究[J];管理工程學(xué)報;2014年02期

5 王潔丹;朱建平;付榮;;函數(shù)型死亡率預(yù)測模型[J];統(tǒng)計研究;2013年09期

6 米子川;趙麗琴;;函數(shù)型數(shù)據(jù)分析的研究進(jìn)展和技術(shù)框架[J];統(tǒng)計與信息論壇;2012年06期

7 李啟華;藍(lán)志青;邢雅康;;基于半?yún)?shù)估計的筆記本電腦特征價格指數(shù)研究[J];東北財經(jīng)大學(xué)學(xué)報;2011年02期

8 黃四民,梁華;用半?yún)?shù)部分線性模型分析居民消費(fèi)結(jié)構(gòu)[J];數(shù)量經(jīng)濟(jì)技術(shù)經(jīng)濟(jì)研究;1994年10期

，

本文編號：1748538