Документ взят из кэша поисковой машины. Адрес оригинального документа : http://hea-www.harvard.edu/AstroStat/astro193/HW11-dueApr27.pdf Дата изменения: Sat Apr 18 21:10:16 2015 Дата индексирования: Sun Apr 10 11:01:45 2016 Кодировка: Поисковые слова: m 103

Astro193 Homework 11
due on April 27, 2015
Periodogram exercise Generate 50 random data points Y(t) assuming a periodic model for the signal: Y (t|P ) = 42.0 + sin (2 t/P ) with a period P = 0.4 day and the time t nonevenly spaced over the total duration time T of: a/ T1= 5 days, b/ T2=10 days c/ T3=30 days. Assume a random Gaussian noise with 0.1 < < 0.5 1/ Plot the data for each observation time T1,T2, T3. 2/ Calculate Nyquist frequency for each observation. 3/ Calculate a periodogram for these data using LombScargle algorithm (you can use the code available in R, IDL or Python SciPy). Plot the periodogram and mark the assumed period on the graph. Do you detect the period P? 4/ Use bootstrap to calculate the significance of the peak. LombScargle periodogram code is available in SciPy, R, MATLAB and IDL.