Python Statistics for Data Science Course- Part 2
Variable and its types
What are the different types of variables?
Confounding variables are those that are not available in the dataset but could have a hidden effect on the results.
In general people who sleep more have better memory and for example we have a data sample with two columns Sleep hours and memory and we are asked to predict, there could be a medical history data that is not available in the data set but could have an impact on the out come. A patient, who has a medical issue related to his memory will have bad memory though he sleeps longer. Here medical condition is a confounding variable that is not available in the data set but could have impact on the out come.
What is a Placebo in a experiment?
A placebo is a non active item(pill) looks same as a active item.
Ex: In a memory test, to differentiate people with more sleep and less sleep, a pill is given to every participant and some pills are placebos which look like a sleeping pill but does not actually have the effect.
Control Variable is that kind of data that can be controlled and reduce the impact.
For example if the results are biased as the samples are collected at the same time, we can reduce this by collecting the values at different time.
Idea is to reduce the no.of confounding variables in the data sample and also reduce the impact of the controlled variables.
What is a Population?
A collection or set of individuals or objects or events whose properties are tend to be analysed is called population.
What is a Sample?
A sample is a subset of population and a well chosen sample will contain most of the information about a particular population parameter.
In general people who sleep more have better memory and for example we have a data sample with two columns Sleep hours and memory and we are asked to predict, there could be a medical history data that is not available in the data set but could have an impact on the out come. A patient, who has a medical issue related to his memory will have bad memory though he sleeps longer. Here medical condition is a confounding variable that is not available in the data set but could have impact on the out come.
What is a Placebo in a experiment?
A placebo is a non active item(pill) looks same as a active item.
Ex: In a memory test, to differentiate people with more sleep and less sleep, a pill is given to every participant and some pills are placebos which look like a sleeping pill but does not actually have the effect.
Control Variable is that kind of data that can be controlled and reduce the impact.
For example if the results are biased as the samples are collected at the same time, we can reduce this by collecting the values at different time.
Idea is to reduce the no.of confounding variables in the data sample and also reduce the impact of the controlled variables.
What is a Population?
A collection or set of individuals or objects or events whose properties are tend to be analysed is called population.
What is a Sample?
A sample is a subset of population and a well chosen sample will contain most of the information about a particular population parameter.
Comments
Post a Comment