SummerStoneS的博客

The bias-variance dilemma (or bias-variance trade-off) is a fundamental concept in machine learning and statistical modeling that describes the trade-off between two types of errors that can occur when building predictive models:Definition: Bias refers to

Cross-validation is a statistical method used to estimate the performance and generalizability of a machine learning model. It involves partitioning the data into subsets, training the model on some of these subsets, and validating the model on the remaini

ConvertUC​1200003600​0.03∣ConvertedPConvertedPConverted∣⋅P​Converted∣VisitedinOctoberUOct​COct​​30000900​0.03PUUOct​​12000030000​∣Converted0.030.03⋅0.25​0.25UOct​COct​​。

假设原假设是对的，观测数据求出的统计量在原假设的分布下的概率，p value是我们能得到比观测值算出的统计量还极端的概率；如果要拒绝原假设，那么alpha至少要比p大，alpha也是一类错误，即原假设是对的，但是拒绝了原假设（本来没效果，说有效果）

The Law of Large Numbers (LLN) and the Central Limit Theorem (CLT) are two fundamental concepts in probability theory and statistics.Law of Large Numbers (LLN):The Law of Large Numbers states that as the size of a sample increases, the sample mean will ge

ARIMA is a popular time series forecasting model that combines three components:ARIMA model parameters are typically chosen through a process called model identification, which involves:Prophet is a forecasting tool developed by Facebook that is designed f

一、数据检查1）缺失值有时候是null,有时候全是0，这个需要比想象中更仔细，因为取数的同学们有时候会默认填充0，有时候是因为数据库迁移，产品刚刚发布所以太远的数据没有等原因造成的，这些需要及时确认对于有些变量仅对特定的人有值，（比如说理财的产品偏好，当然只有买了理财的人才会有偏好数据可以分析），可以给没有这个值的人填充一个数值，可以是999这样这个字段不会取到的大值（但是要注意在模型...