开荒暗夜不顺,这锅我背了,各种死(就仨奶,死一个肯定灭啊)
7.2 之前肝出 各种大特质
培养学习好习惯,早起先干不开心的事
1. describe a set of data
2. find relationships I don't know about ' correlation does not imply causation, 不是最终解,不能只用它来预测
3. use small sample of data to say sth about a bigger population (目标,包含总量和置信度)
4. ro use the data on some objects to predict values for another object
5. to find out what happens to one variable when you make another variable change(随机试验需要分析因果,不随机的对影响因子很敏感,结果一般显示为平均值 不能代表个体)
6. understand the exact changes in variables that lead to changes in other variables for individual objects
definition of data
data are values of qualitative or quantitative variables, belonging to a set of items.
最重要的是问题
其次的才是数据
big data
Experimental Design:
* result have replication, measure variability
* trouble
* analysis plan: generalize to the problem you care about
* share data and experience is transparent
Design my experiment
* formulate my question in advance
* confounding
* correlation inference (easy)
* randomization and blocking
-fix a variable
-stratify
-randomize observe the outcome
* prediction ‘is not inference
-quantities: + /-
* data dredging 钓鱼科研 手动滑稽
tackling procrastination
* keep a planner journal
* commit yourself to certain routines and tasks each day
* delay rewards until you finish the task
* watch for procrastination cue
* gain trust in your new system
* have backup plans for when you still procrastinate
* eat your frogs first