Write a Python function to calculate various descriptive statistics metrics for a given dataset. The function should take a list or NumPy array of numerical values and return a dictionary containing:
- mean: Average of all values
- median: Middle value when sorted
- mode: Most frequently occurring value, returned as an
int. If several values tie, return the smallest of them (so for all-distinct data it is the minimum)
- variance: Population variance (divide by N)
- standard_deviation: Square root of variance
- 25th_percentile, 50th_percentile, 75th_percentile: Quartile values, using linear interpolation between the sorted values (the default of
np.percentile)
- interquartile_range: Difference between 75th and 25th percentiles (IQR)
Every value except the mode is a float. Keep the keys in the order listed above.