1 / 12

Fitting (special modeling)

Fitting (special modeling). 预习 BR2003, Chap. 6. 董小波 2009.11.18. Schematic. The role of data analysis. On the learning of curve fitting. 分清三种性质的学习内容: 原理性 技术性 / 方法性 操作性 由于选课同学层次弥散极大,请各人根据不同的阶段、个人不同的需求等,给予不同的学习时间权重。 在 独观大略 与 惟务精纯 之间平衡. 3. 想象力 与 数据分析 之间的平衡.

minnie
Download Presentation

Fitting (special modeling)

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Fitting (special modeling) 预习BR2003, Chap. 6 董小波 2009.11.18

  2. Schematic The role of data analysis

  3. On the learning of curve fitting 分清三种性质的学习内容: • 原理性 • 技术性/方法性 • 操作性 • 由于选课同学层次弥散极大,请各人根据不同的阶段、个人不同的需求等,给予不同的学习时间权重。 • 在独观大略 与 惟务精纯 之间平衡

  4. 3. 想象力 与 数据分析 之间的平衡

  5. Anyway, data are necessary. Data! Data! He cried impatiently, I can’t make bricks without clay. --- <The Copper Beeches>, Conan Doyle Then, The science and art of data handling and decision making: Data Analysis

  6. 数学与统计 统计的理论基础: 现代科学视野下的偶然性与因果律 量子力学的几率解释,不确定性原理; 物理定律(如相对论、量子力学等)的变换不变性 [若干例子。人(观察者)有一定的自由,然而不昧因果。] 变量vs.随机变量 知识:对不确定性的量度 —— “给出概率!” 。 一种(新的)思维方法:基于概率的推理方法。 BTW, even a possibility:因果律是一定初始条件(或选择)下的 大数极限(统计学近似)。

  7. 进一步的问题 偶然性(随机、不确定性等)的起源? E.g., 测量误差中,量子涨落(then, what is the origin of the quantum uncertainty? Unknown to us still.) 另一种可能:随机性来源于底层的确定论系统,是一种(极好的)统计近似性质。比如布朗运动: Anyway, what we concern in this course: 应用随机性,应用统计学理论与方法,从而解决天体物理中的问题。

  8. Curve fitting • Step 1: 选定函数模型(假设/假说) • Step 2: 模型是否正确?(假说检验) 【e.g., 参数个数是否足够?】 • Step 3: 拟合得到的最佳参数值,及其置信区间 Our goal: Grasp and practice general data analysis in one semester (60hrs).

  9. Overview of Chap.6 • Focused on Subsection 6.2 specifically, careful only for ``Method of Maximum Likelihood’’. • To glance over Subsection 6.3 • To read ``Chisq Probability’’ (p.108) and ``Uncertanties in the Parameters’’ (p.109) [Note: actually, nobody bothers to estimate the Uncertanties in the Parameters using the formal error propagation equation.] • To skip other sections if you have no time/interest.

  10. Only one thing in this Chapter you should understand z Maximizing the likelyhood  minimizing 统计量z (在最佳参数值a,b附近,z符合chi^2分布。后面讲述。)

  11. For the next class • Read in advance chapter 7. • Pay much time to finish the case, see: ftp://210.45.66.48/teaching/methods09/Course_Notes_and_Homeworks/fitting/ or http://ustcastroph.blog.sohu.com/#tp_95a1b1f5a7a Advice: practice is more important than reading.

  12. Case discussion Today 现有一个类星体光谱观测样本,红移在0.45—0.8之间,共2092个源。根据光谱,我们可以测得MgII λ2800A发射线的半高全宽(FWHM)和等值宽度(EW);可以测得类星体连续谱在3000A处的光度[ L3000= 3000A * L_lambda(3000) ]。根据以上参数我们还可以计算出类星体中心黑洞的质量(M)和Eddington ratio(L/Ledd)。数据及其简要说明见data_for_spearman.txt文件。 我们要研究:EW(MgII)大小是否由以上某一个参量主要决定(L3000, FWHM, M, L/Ledd),为此我们开展相关与偏相关分析。最后,根据Spearman相关与偏相关分析的结果,给出你的结论。

More Related