A hidden Markov model for estimating age‐specific survival when age and size are uncertain

Abstract

Estimates of age-specific survival probabilities are needed for age-structured population models and to inform conservation decisions. However, determining the age of individuals in wildlife populations is often problematic. We present a hidden Markov model for estimating age-specific survival from capture-recapture or capture-recapture-recovery data when age is unknown and indicators of age, such as size and growth layer counts, are imprecise. The model is evaluated through simulations, and its implementation is illustrated with maximum likelihood and Bayesian approaches in commonly used software. The model is then applied to genetic capture-recapture data of Florida manatees to estimate age-and time-variant survival probabilities. The approach is broadly applicable to studies aiming to quantify age-specific effects of environmental change and management actions on population dynamics, including studies that rely on minimally invasive methods such as genetic-and photo-identification.

Application of Hidden Markov Model in Financial Time Series Data

“…erefore, there must be a lot of information about economic laws hidden in the financial time series. Finding all kinds of valuable information from financial time series and better understanding its application is undoubtedly of great significance to investment forecasting, decision making, and risk management [1][2][3][4][5][6][7][8][9].…”