1/n To achieve something big we need to start with taking small steps. It's all about thinking from the first principles and making our foundations strong. To make into the field of data science one of the stepping stones is learning and understanding statistics.
1/n Here we saw that covariance has certain disadvantages, to overcome those we can use another statistic called Correlation to measure the strength of relationship between variables
https://t.co/gLyiOJBWRX
9/n Covariance has weird units which don't make much sense as seen earlier with cm-lbs. And for these reasons Covariance is difficult to interpret. Next thread we will see Correlation
I am planning to attend a bunch of music fests this year. Yayy excited. But my single biggest concern is finding the cleanest toilet. Turns out, I finally have an optimal solution for my success in this quest. Wanna know how? Follow the thread!
So yes, where does the k/k+1 come from. If k+2 bathroom is the best option, to choose this, I must not choose k+1st bathroom. After solving for x, we reach at k and P(s) both being equal to 37.
See below:
6/n A confidence interval (CI) provides a range of values that are included in a given fraction of a sampling distribution. 90% CI would be a range of values between the 5th and 95th percentile. It does NOT mean there is 90% probability of population parameter being in that range
1/n Continuing the discussion of estimation and estimators from the thread https://t.co/nO1qvKZHKK
When a value is being calculated for the entire population it is called a parameter and the corresponding term for a sample is called a statistic.
5/n Standard error is the standard deviation in the sampling distribution of the sample means. Standard error measures, on average, how much do we expect the estimate to deviate from the true value. In turn, it's nothing but RMSE as seen here
https://t.co/i6zoCQ6J1P
11/n In the absence of an outlier, this sample mean 'xbar' minimizes the Mean Squared Error (MSE) = Summation of all the values of (xbar - u)/m, where 'm' is the number iterations. RMSE is nothing but the Square root of MSE
1/n Continuing the discussion of estimation and estimators from the thread https://t.co/nO1qvKZHKK
When a value is being calculated for the entire population it is called a parameter and the corresponding term for a sample is called a statistic.
1/n In this tweet thread, I talk about what statistics is. In nutshell, it is the study of gathering data and extrapolating information regarding the population from a sample which is called Estimation.
https://t.co/tFM7Ws4q3U
#stats#Statistics#estimation
6/n The above formula doesn't tend to underestimate the variance for a larger sample and also yields a bias of Zero as seen in the wiki link: https://t.co/USjNJrD9Rr