Expansion computed for the quadratic map for 1025 parameters using dynamically refined partitions with different radii of the critical neighborhood

Expansion data computed for the quadratic map by the program that implements the algorithms introduced in the paper “Rigorous computation of expansion in one-dimensional dynamics” by Paweł Pilarczyk, Michał Palczewski and Stefano Luzzatto.
This computation was conducted for 1025 uniformly spaced parameter values in [1.4,2] using dynamically refined partitions of up to k intervals outside the critical neighborhood of radius δ. There are four datasets in the package obtained for a few different values of k and δ:– run20e00.csv: δ=0.001, k=3000– run20e01.csv: δ=0.1, k=1000– run20e02.csv: δ=0.01, k=1000– run20e03.csv: δ=0.001, k=1000
The data is in the CSV format, with the first row containing column labels. The contents of the columns is the following:

level — the level of subdivision of the parameter interval (e.g. 10 for 2^10=1024 subintervals)
num — the identifier of the data piece in the collection at the given subdivision; the identifiers begin with 0
parMin — the left endpoint of the parameter interval (minimal parameter value)
parMax — the right endpoint of the parameter interval (maximal parameter value)
k — the total number of intervals on which the graph representation of the map was built (the critical neighborhood is counted here, too)
delta — the radius δ of the critical neighborhood
lambda — the computed expansion exponent λ
logC — log C if the constant C was computed, otherwise 0
lambda0 — the constant λ₀ if it was computed, otherwise 0
period — the period of a periodic orbit found (0 if none)
lambdaMax — an upper bound on the expansion exponent of the periodic orbit found (0 if none)
distFrom0out — the minimum guaranteed distance of the periodic orbit from 0
distFrom0in — an upper bound on the distance from 0 during the closest approach to 0
compTime — the computation time measured in seconds

This research was supported by the National Science Centre, Poland, within the grant OPUS 2021/41/B/ST1/00405. Some computations were carried out at the Centre of Informatics Tricity Academic Supercomputer & Network.