A team of researchers has already formulated their hypotheses and now they want to test these against their collected data. What type of data analysis would be appropriate?
- All are equally suitable
- CDA
- EDA
- Predictive Modeling
CDA would be the most appropriate as it involves testing pre-formulated hypotheses against the collected data to either confirm or refute them.
You've created a pairplot of your dataset, and one scatter plot in the grid shows a clear linear pattern. What could this potentially indicate?
- The two variables are highly uncorrelated
- The two variables are unrelated
- The two variables have a strong linear relationship
- The two variables have no relationship
If a scatter plot in a pairplot shows a clear linear pattern, this could potentially indicate that the two variables have a strong linear relationship. This means that changes in one variable correspond directly to changes in the other variable.
Why might PCA be considered a method of feature selection?
- It can handle correlated features
- It can improve model performance
- It can reduce the dimensionality of the data
- It transforms the data into a new space
Principal Component Analysis (PCA) can be considered a method of feature selection because it reduces the dimensionality of the data by transforming the original features into a new set of uncorrelated features. These new features, called principal components, are linear combinations of the original features and are selected to capture the most variance in the data.
In regression analysis, if the Variance Inflation Factor (VIF) for a predictor is 1, this means that _________.
- The predictor is not at all correlated with other predictors
- The predictor is not at all correlated with the response
- The predictor is perfectly correlated with other predictors
- The predictor is perfectly correlated with the response
In regression analysis, a Variance Inflation Factor (VIF) of 1 indicates that there is no correlation between the given predictor and the other predictors. This implies no multicollinearity.
When performing a pairwise analysis, _____ deletion discards only the specific pairs of data where one is missing.
- Listwise
- Pairwise
- Random
- Systematic
When performing a pairwise analysis, 'pairwise' deletion discards only the specific pairs of data where one is missing. It allows the retention of more data compared to listwise deletion, but it can lead to biased results if the data is not missing completely at random.
You have found that your dataset has a high degree of multicollinearity. What steps would you consider to rectify this issue?
- Add more data points
- Increase the model bias
- Increase the model complexity
- Use Principal Component Analysis (PCA)
One way to rectify multicollinearity is to use Principal Component Analysis (PCA). PCA transforms the original variables into a new set of uncorrelated variables, thereby removing multicollinearity.
How can histograms be used to detect outliers?
- Outliers are represented by bars that are far away from others
- Outliers are represented by the shortest bars
- Outliers are represented by the tallest bars
- Outliers cannot be detected with histograms
In a histogram, outliers can often be represented by bars that are noticeably separated from the rest of the data distribution.
What is a correlation matrix and what is its primary purpose in Exploratory Data Analysis?
- A graphical representation of the correlation between variables
- A representation of missing values in the data
- A representation of the data distribution
- A visual representation of data clusters
A correlation matrix is a tabular data representing the correlations between pairs of variables. Each cell in the table shows the correlation between two variables. It's primary use in EDA is to understand the linear relationship between the variables.
Regularization techniques like Ridge and Lasso can indirectly perform feature selection by assigning a _______ coefficient to irrelevant features.
- Negative
- Non-zero
- Positive
- Zero
Regularization techniques like Ridge and Lasso can indirectly perform feature selection by assigning a zero coefficient to irrelevant features. This is achieved by adding a penalty term to the loss function that encourages smaller or zero coefficients, effectively removing the irrelevant features from the model.
What type of data is Spearman's correlation most suitable for?
- Categorical data
- Continuous, normally distributed data
- Nominal data
- Ordinal data
Spearman's correlation is most suitable for ordinal data. It assesses how well the relationship between two variables can be described using a monotonic function. Because it's based on ranks, it can be used with ordinal data, where the order is important but not the difference between values.
Modified Z-score is a more robust estimator in the presence of _______.
- normally distributed data
- outliers
- skewed data
- uniformly distributed data
The modified Z-score is more robust in the presence of outliers, making it better suited to datasets with many extreme values.
How does multicollinearity affect feature selection?
- It affects the accuracy of the model
- It causes unstable parameter estimates
- It makes the model less interpretable
- It results in high variance of the model
Multicollinearity, which refers to the high correlation between predictor variables, can affect feature selection by causing unstable estimates of the parameters. This instability can lead to strange and unreliable predictions, making the feature selection process less accurate.