x*********0
发帖数: 651 | 1 我太菜了，听对方的口气就能感觉出来。唉
下面是问到的题目，大家看看，正确答案应该是什么，就当学习了。
1. 100 records, 1 response variable, 100+predictor. x1 is the most important
variable that should be used in the model. However, when y vs. x1 plot, no
pattern can be found. whY?
My anwer: grouping
2. How to select variables (from 1):
My answer: stepwise will be considered. We can also use best subset, but the
predictor is too much, I don’t know if it is applicable.
3. How to evaluate a model:
My answer: adj R square, F, Multicollinearity
4. if | l********s
发帖数: 88 | 2 问问lz这是什么类型的公司什么类型的职位啊？
纯属好奇，什么样的职位电面会问这么专业的统计知识呢？我碰到的都没有问这么专业
的。。。
important
no
the
【在 x*********0 的大作中提到】
: 我太菜了，听对方的口气就能感觉出来。唉
: 下面是问到的题目，大家看看，正确答案应该是什么，就当学习了。
: 1. 100 records, 1 response variable, 100+predictor. x1 is the most important
: variable that should be used in the model. However, when y vs. x1 plot, no
: pattern can be found. whY?
: My anwer: grouping
: 2. How to select variables (from 1):
: My answer: stepwise will be considered. We can also use best subset, but the
: predictor is too much, I don’t know if it is applicable.
: 3. How to evaluate a model:
| x*********0
发帖数: 651 | 3 本地的一个小公司，是一个纯技术人员面试的，什么客套都没有，上来就问。职位就是
statistician | d******e
发帖数: 7844 | 4 你这是什么公司啊？居然考这些问题？
1. grouping是什么概念？第一题不就是说明X1不足以解释Y的么？
2. 赶时髦的话可以试试Lasso, Elastic net或者其他shrinkage methods?
3. BIC,AIC,Cross Validation,Generalized Degree of Freedom, Convariance
Penalty等等?
5. 这个老师上课说过，一般是把所有的marginal的variable都算进去，因为漏掉一个
variable比多一个冗余variable对model的影响要大得多。
important
no
the
should
【在 x*********0 的大作中提到】
: 我太菜了，听对方的口气就能感觉出来。唉
: 下面是问到的题目，大家看看，正确答案应该是什么，就当学习了。
: 1. 100 records, 1 response variable, 100+predictor. x1 is the most important
: variable that should be used in the model. However, when y vs. x1 plot, no
: pattern can be found. whY?
: My anwer: grouping
: 2. How to select variables (from 1):
: My answer: stepwise will be considered. We can also use best subset, but the
: predictor is too much, I don’t know if it is applicable.
: 3. How to evaluate a model:
| b*******r
发帖数: 152 | 5 food for thought--
1. grouping/subgrouping/segmentation is a good answer. another direction is
to do the transform of x1, eg. log(x1).
2. i think he is looking for some variable reduction techniques, like
clustering, pca, fa, etc...
3.4.i think they are more likely open questions.
5.can u say say why is yr answer?
thx | b**********i
发帖数: 1059 | | S******o
发帖数: 68 | | s*r
发帖数: 2757 | 8 using F test to judge model?
adding more x variables in the regression to reduce mulcollinearity?
important
no
the
【在 x*********0 的大作中提到】
: 我太菜了，听对方的口气就能感觉出来。唉
: 下面是问到的题目，大家看看，正确答案应该是什么，就当学习了。
: 1. 100 records, 1 response variable, 100+predictor. x1 is the most important
: variable that should be used in the model. However, when y vs. x1 plot, no
: pattern can be found. whY?
: My anwer: grouping
: 2. How to select variables (from 1):
: My answer: stepwise will be considered. We can also use best subset, but the
: predictor is too much, I don’t know if it is applicable.
: 3. How to evaluate a model:
| c*********n
发帖数: 87 | 9 R is linear correlation coefficient which is used to evaluate the linear
relationship between the dependent and independent variables. If the model
is a linear model, then R^2 means the model explains 30% of variability of
the dependent variable. If the model is not a linear model, then R^2 could
not make inference for the model validation. | s*r
发帖数: 2757 | 10 r2 can be used in nonlinear situations
【在 c*********n 的大作中提到】
: R is linear correlation coefficient which is used to evaluate the linear
: relationship between the dependent and independent variables. If the model
: is a linear model, then R^2 means the model explains 30% of variability of
: the dependent variable. If the model is not a linear model, then R^2 could
: not make inference for the model validation.
| h***x
发帖数: 586 | 11 都是实际中常用的问题
important
no
the
【在 x*********0 的大作中提到】
: 我太菜了，听对方的口气就能感觉出来。唉
: 下面是问到的题目，大家看看，正确答案应该是什么，就当学习了。
: 1. 100 records, 1 response variable, 100+predictor. x1 is the most important
: variable that should be used in the model. However, when y vs. x1 plot, no
: pattern can be found. whY?
: My anwer: grouping
: 2. How to select variables (from 1):
: My answer: stepwise will be considered. We can also use best subset, but the
: predictor is too much, I don’t know if it is applicable.
: 3. How to evaluate a model:
| b********n
发帖数: 95 | 12 why you think that could be a good model when R2 around 30%? That should be
pretty bad result for correlation ba.
important
no
the
【在 x*********0 的大作中提到】
: 我太菜了，听对方的口气就能感觉出来。唉
: 下面是问到的题目，大家看看，正确答案应该是什么，就当学习了。
: 1. 100 records, 1 response variable, 100+predictor. x1 is the most important
: variable that should be used in the model. However, when y vs. x1 plot, no
: pattern can be found. whY?
: My anwer: grouping
: 2. How to select variables (from 1):
: My answer: stepwise will be considered. We can also use best subset, but the
: predictor is too much, I don’t know if it is applicable.
: 3. How to evaluate a model:
| b********n
发帖数: 95 | 13 and question 5, you should at least ask about the p-value for each model ba.
.....
maybe one of x3 or x4 is just some useless variable.
important
no
the
【在 x*********0 的大作中提到】
: 我太菜了，听对方的口气就能感觉出来。唉
: 下面是问到的题目，大家看看，正确答案应该是什么，就当学习了。
: 1. 100 records, 1 response variable, 100+predictor. x1 is the most important
: variable that should be used in the model. However, when y vs. x1 plot, no
: pattern can be found. whY?
: My anwer: grouping
: 2. How to select variables (from 1):
: My answer: stepwise will be considered. We can also use best subset, but the
: predictor is too much, I don’t know if it is applicable.
: 3. How to evaluate a model:
|
|
