Bayesian analysis of spherically parameterized dynamic multivariate stochastic volatility models

Collaborative analysis for energy usage monitoring and management on a large university campus

“…We discussed (a) in-depth in Section 4; this information transfer occurs on a weekly basis from consultant to client. Two publications (Hu et al (2022) and Lim et al (2022)) and an arXiv article (Linder et al, 2021)…”

“…We discussed (a) in‐depth in Section 4; this information transfer occurs on a weekly basis from consultant to client. Two publications (Hu et al (2022) and Lim et al (2022)) and an arXiv article (Linder et al, 2021) have resulted from this collaboration thus far. Three undergraduate students (Yutong Chen, Ziyi Kang, and Hang Zeng) presented posters in the 22nd Annual Frontiers in Undergraduate Research, at UConn.…”

Hierarchical modeling of irregularly spaced financial returns

“…We assume that $$$ \mu \sim N\left(0,10\right),{\sigma}_{\eta}^2\sim IG\left(1,1\right),\frac{\phi +1}{2}\sim \mathrm{Beta}\left(1,1\right),\omega \sim \mathrm{Gamma}\left(\mathrm{0.001,0.001}\right),\delta \sim \mathrm{Gamma}\left(\mathrm{0.001,0.001}\right) $$$, $$$ \left({\alpha}_1,\dots, {\alpha}_p,{\beta}_1,\dots, {\beta}_q,{\beta}_{q+1}\right)\sim \mathrm{Dirichlet}\left(0.5,\dots, 0.5\right) $$$. Priors for $$$ \mu, \phi $$$ and $$$ {\sigma}_{\eta}^2 $$$ have been set up similar to Hu et al (2023). We assign noninformative gamma priors to both $$$ \omega $$$ and $$$ \delta $$$, assuming positive values with mean 1 and variance 1000.…”

“…We assume that μ $ Nð0,10Þ, σ 2 η $ IGð1,1Þ, ϕþ1 2 $ Betað1,1Þ, ω $ Gammað0:001,0:001Þ, δ $ Gammað0:001,0:001Þ, ðα 1 , …, α p , β 1 ,…, β q ,β qþ1 Þ $ Dirichletð0:5,…,0:5Þ. Priors for μ, ϕ and σ 2 η have been set up similar to Hu et al (2023). We assign noninformative gamma priors to both ω and δ, assuming positive values with mean 1 and variance 1000.…”