a question on transformation - Statistics版

本页内容为未名空间相应帖子的节选和存档，一周内的贴子最多显示50字，超过一周显示500字访问原贴

Statistics版 - a question on transformation

相关主题
● How to transform predictor variable?	● 请问各位能人一个data transformation的问题
● R classification tree model 请教	● 一般什么样的数据，分析起来要做log transformation?
● support vector machine	● transformation
● 请问一个Regression的问题	● mining time series data (转载)
● 线性回归中log tranformation	● Re: help, how to make a data normal?
● y = a + b * (c ^ x)	● [合集] 纺锤型的残差图代表什么意思？
● [sas] 怎么用自定义format里的值来计算？	● what is statistic score
● 求助：为什么box-cox transformed 后的varibale 不normal	● [合集] linear regression的问题

相关话题的讨论汇总
话题: question话题: flat话题: up话题: down

进入Statistics版参与讨论

(共1页)

e****d
发帖数: 333

a variable/predictor, x_t, indexed by t.
it can be flat for a period of time, then up, then down, up, down, etc.
is there a way to transform this x_t to a y_t, so that in a downtrend, y_t
is much lower, but when in a flat/up trend, y_t is roughly the same as x_t?
Thanks.

(共1页)

进入Statistics版参与讨论

相关主题
● [合集] linear regression的问题	● 线性回归中log tranformation
● 发包子请人review我的syllabus - stochastics	● y = a + b * (c ^ x)
● [合集] how to prove this problem ?	● [sas] 怎么用自定义format里的值来计算？
● 紧急求助: R heatmap	● 求助：为什么box-cox transformed 后的varibale 不normal
● How to transform predictor variable?	● 请问各位能人一个data transformation的问题
● R classification tree model 请教	● 一般什么样的数据，分析起来要做log transformation?
● support vector machine	● transformation
● 请问一个Regression的问题	● mining time series data (转载)

相关话题的讨论汇总
话题: question话题: flat话题: up话题: down

#	版面	帖数(主题数)
-	全站	4871 (796)
1	Military	3777 (569)
2	Stock	341 (51)
3	Joke	117 (17)
4	History	116 (3)
5	Automobile	100 (9)
6	USANews	55 (9)
7	Midlife	45 (1)
8	Headline	41 (41)
9	Dreamer	33 (13)
10	FleaMarket	32 (20)
11	Living	30 (7)

boards

未名新帖统计// 7月16日

历史上的今天