tag:crantastic.org,2005:/authors/4718Latest activity for Hyungsuk Tak2018-05-16T17:03:11Zcrantastic.orgtag:crantastic.org,2005:TimelineEvent/839462019-01-23T06:20:32Z2019-01-23T06:20:32Zcarfima was upgraded to version 2.0.1<a href="/packages/carfima">carfima</a> was <span class="action">upgraded</span> to version <a href="/packages/carfima/versions/79810">2.0.1</a><br /><h3>Package description:</h3><p>We provide a toolbox to fit a continuous-time fractionally integrated ARMA process (CARFIMA) on univariate and irregularly spaced time series data via frequentist or Bayesian machinery. A general-order CARFIMA(p, H, q) model for p>q is specified in Tsai and Chan (2005) <doi:10.1111/j.1467-9868.2005.00522.x> and it involves (p+q+2) unknown model parameters, i.e., p AR parameters, q MA parameters, Hurst parameter H, and process uncertainty (standard deviation) sigma. The package produces their maximum likelihood estimates and asymptotic uncertainties using a global optimizer called the differential evolution algorithm. It also produces their posterior distributions via Metropolis within a Gibbs sampler equipped with adaptive Markov chain Monte Carlo for posterior sampling. These fitting procedures, however, may produce numerical errors if p>2. The toolbox also contains a function to simulate discrete time series data from CARFIMA(p, H, q) process given the model parameters and observation times.</p>crantastic.orgtag:crantastic.org,2005:TimelineEvent/812802018-11-01T22:00:34Z2018-11-01T22:00:34Zcarfima was upgraded to version 2.0.0<a href="/packages/carfima">carfima</a> was <span class="action">upgraded</span> to version <a href="/packages/carfima/versions/77371">2.0.0</a><br /><h3>Package description:</h3><p>We provide a toolbox to fit a continuous-time fractionally integrated ARMA process (CARFIMA) on univariate and irregularly spaced time series data via frequentist or Bayesian machinery. A general-order CARFIMA(p, H, q) model for p>q is specified in Tsai and Chan (2005) <doi:10.1111/j.1467-9868.2005.00522.x> and it involves (p+q+2) unknown model parameters, i.e., p AR parameters, q MA parameters, Hurst parameter H, and process uncertainty (standard deviation) sigma. The package produces their maximum likelihood estimates and asymptotic uncertainties using a global optimizer called the differential evolution algorithm. It also produces their posterior distributions via Metropolis within a Gibbs sampler equipped with adaptive Markov chain Monte Carlo for posterior sampling. These fitting procedures, however, may produce numerical errors if p>2. The toolbox also contains a function to simulate discrete time series data from CARFIMA(p, H, q) process given the model parameters and observation times.</p>crantastic.orgtag:crantastic.org,2005:TimelineEvent/750892018-05-16T17:03:11Z2018-05-16T17:03:11Ztimedelay was upgraded to version 1.0.8<a href="/packages/timedelay">timedelay</a> was <span class="action">upgraded</span> to version <a href="/packages/timedelay/versions/71691">1.0.8</a><br /><h3>Package description:</h3><p>We provide a toolbox to estimate the time delay between the brightness time series of gravitationally lensed quasar images via Bayesian and profile likelihood approaches. The model is based on a state-space representation for irregularly observed time series data generated from a latent continuous-time Ornstein-Uhlenbeck process. Our Bayesian method adopts scientifically motivated hyper-prior distributions and a Metropolis-Hastings within Gibbs sampler, producing posterior samples of the model parameters that include the time delay. A profile likelihood of the time delay is a simple approximation to the marginal posterior distribution of the time delay. Both Bayesian and profile likelihood approaches complement each other, producing almost identical results; the Bayesian way is more principled but the profile likelihood is easier to implement.</p>crantastic.orgtag:crantastic.org,2005:TimelineEvent/693532017-12-11T12:15:31Z2017-12-11T12:15:31Zcrantastic_production tagged carfima with TimeSeries<a href="/users/146">crantastic_production</a> <span class="action">tagged</span> <a href="/packages/carfima">carfima</a> with <a href="/task_views/TimeSeries">TimeSeries</a>crantastic_productiontag:crantastic.org,2005:TimelineEvent/676302017-10-23T09:20:21Z2017-10-23T09:20:21Zcarfima was released<a href="/packages/carfima">carfima</a> was <span class="action">released</span><br /><h3>Package description:</h3><p>We provide a toolbox to fit a continuous-time fractionally integrated ARMA process (CARFIMA) on univariate and irregularly spaced time series data via frequentist or Bayesian machinery. A general-order CARFIMA(p, H, q) model for p>q is specified in Tsai and Chan (2005) <doi:10.1111/j.1467-9868.2005.00522.x> and it involves (p+q+2) unknown model parameters, i.e., p AR parameters, q MA parameters, Hurst parameter H, and process uncertainty (standard deviation) sigma. The package produces their maximum likelihood estimates and asymptotic uncertainties using a global optimizer called the differential evolution algorithm. It also produces their posterior distributions via Metropolis within a Gibbs sampler equipped with adaptive Markov chain Monte Carlo for posterior sampling. These fitting procedures, however, may produce numerical errors if p>2. The toolbox also contains a function to simulate discrete time series data from CARFIMA(p, H, q) process given the model parameters and observation times.</p>crantastic.orgtag:crantastic.org,2005:TimelineEvent/629792017-05-27T21:42:20Z2017-05-27T21:42:20Ztimedelay was upgraded to version 1.0.7<a href="/packages/timedelay">timedelay</a> was <span class="action">upgraded</span> to version <a href="/packages/timedelay/versions/60226">1.0.7</a><br /><h3>Package description:</h3><p>We provide a toolbox to estimate the time delay between the brightness time series of gravitationally lensed quasar images via Bayesian and profile likelihood approaches. The model is based on a state-space representation for irregularly observed time series data generated from a latent continuous-time Ornstein-Uhlenbeck process. Our Bayesian method adopts scientifically motivated hyper-prior distributions and a Metropolis-Hastings within Gibbs sampler, producing posterior samples of the model parameters that include the time delay. A profile likelihood of the time delay is a simple approximation to the marginal posterior distribution of the time delay. Both Bayesian and profile likelihood approaches complement each other, producing almost identical results; the Bayesian way is more principled but the profile likelihood is easier to implement.</p>crantastic.orgtag:crantastic.org,2005:TimelineEvent/606452017-03-23T06:41:50Z2017-03-23T06:41:50Ztimedelay was upgraded to version 1.0.6<a href="/packages/timedelay">timedelay</a> was <span class="action">upgraded</span> to version <a href="/packages/timedelay/versions/58044">1.0.6</a><br /><h3>Package description:</h3><p>We provide a toolbox to estimate the time delay between the brightness time series of gravitationally lensed quasar images via Bayesian and profile likelihood approaches. The model is based on a state-space representation for irregularly observed time series data generated from a latent continuous-time Ornstein-Uhlenbeck process. Our Bayesian method adopts scientifically motivated hyper-prior distributions and a Metropolis-Hastings within Gibbs sampler, producing posterior samples of the model parameters that include the time delay. A profile likelihood of the time delay is a simple approximation to the marginal posterior distribution of the time delay. Both Bayesian and profile likelihood approaches complement each other, producing almost identical results; the Bayesian way is more principled but the profile likelihood is easier to implement.</p>crantastic.orgtag:crantastic.org,2005:TimelineEvent/576302016-12-13T07:22:09Z2016-12-13T07:22:09Ztimedelay was upgraded to version 1.0.5<a href="/packages/timedelay">timedelay</a> was <span class="action">upgraded</span> to version <a href="/packages/timedelay/versions/55181">1.0.5</a><br /><h3>Package description:</h3><p>We provide a toolbox to estimate the time delay between the brightness time series of gravitationally lensed quasar images via Bayesian and profile likelihood approaches. The model is based on a state-space representation for irregularly observed time series data generated from a latent continuous-time Ornstein-Uhlenbeck process. Our Bayesian method adopts scientifically motivated hyper-prior distributions and a Metropolis-Hastings within Gibbs sampler, producing posterior samples of the model parameters that include the time delay. A profile likelihood of the time delay is a simple approximation to the marginal posterior distribution of the time delay. Both Bayesian and profile likelihood approaches complement each other, producing almost identical results; the Bayesian way is more principled but the profile likelihood is easier to implement.</p>crantastic.orgtag:crantastic.org,2005:TimelineEvent/572012016-11-30T01:21:48Z2016-11-30T01:21:48Ztimedelay was upgraded to version 1.0.4<a href="/packages/timedelay">timedelay</a> was <span class="action">upgraded</span> to version <a href="/packages/timedelay/versions/54800">1.0.4</a><br /><h3>Package description:</h3><p>We provide a toolbox to estimate the time delay between the brightness time series of gravitationally lensed quasar images via Bayesian and profile likelihood approaches. The model is based on a state-space representation for irregularly observed time series data generated from a latent continuous-time Ornstein-Uhlenbeck process. Our Bayesian method adopts scientifically motivated hyper-prior distributions and a Metropolis-Hastings within Gibbs sampler, producing posterior samples of the model parameters that include the time delay. A profile likelihood of the time delay is a simple approximation to the marginal posterior distribution of the time delay. Both Bayesian and profile likelihood approaches complement each other, producing almost identical results; the Bayesian way is more principled but the profile likelihood is easier to implement.</p>crantastic.orgtag:crantastic.org,2005:TimelineEvent/564632016-11-04T21:41:52Z2016-11-04T21:41:52Ztimedelay was upgraded to version 1.0.3<a href="/packages/timedelay">timedelay</a> was <span class="action">upgraded</span> to version <a href="/packages/timedelay/versions/54116">1.0.3</a><br /><h3>Package description:</h3><p>We provide a toolbox to estimate the time delay between the brightness time series of gravitationally lensed quasar images via Bayesian and profile likelihood approaches. The model is based on a state-space representation for irregularly observed time series data generated from a latent continuous-time Ornstein-Uhlenbeck process. Our Bayesian method adopts scientifically motivated hyper-prior distributions and a Metropolis-Hastings within Gibbs sampler, producing posterior samples of the model parameters that include the time delay. A profile likelihood of the time delay is a simple approximation to the marginal posterior distribution of the time delay. Both Bayesian and profile likelihood approaches complement each other, producing almost identical results; the Bayesian way is more principled but the profile likelihood is easier to implement.</p>crantastic.orgtag:crantastic.org,2005:TimelineEvent/472392016-02-11T16:12:54Z2016-02-11T16:12:54Ztimedelay was upgraded to version 1.0.2<a href="/packages/timedelay">timedelay</a> was <span class="action">upgraded</span> to version <a href="/packages/timedelay/versions/46996">1.0.2</a><br /><h3>Package description:</h3><p>We provide a toolbox to estimate the time delay between the brightness time series of gravitationally lensed quasar images via Bayesian and profile likelihood approaches. The model is based on a state-space representation for irregularly observed time series data generated from a latent continuous-time Ornstein-Uhlenbeck process. Our Bayesian method adopts scientifically motivated hyper-prior distributions and a Metropolis-Hastings within Gibbs sampler, producing posterior samples of the model parameters that include the time delay. A profile likelihood of the time delay is a simple approximation to the marginal posterior distribution of the time delay. Both Bayesian and profile likelihood approaches complement each other, producing almost identical results; the Bayesian way is more principled but the profile likelihood is easier to implement.</p>crantastic.orgtag:crantastic.org,2005:TimelineEvent/462802016-01-05T13:32:47Z2016-01-05T13:32:47Ztimedelay was upgraded to version 1.0.1<a href="/packages/timedelay">timedelay</a> was <span class="action">upgraded</span> to version <a href="/packages/timedelay/versions/46051">1.0.1</a><br /><h3>Package description:</h3><p>We provide a toolbox to estimate the time delay between the brightness time series of gravitationally lensed quasar images via Bayesian and profile likelihood approaches. The model is based on a state-space representation for irregularly observed time series data generated from a latent continuous-time Ornstein-Uhlenbeck process. Our Bayesian method adopts scientifically motivated hyper-prior distributions and a Metropolis-Hastings within Gibbs sampler, producing posterior samples of the model parameters that include the time delay. A profile likelihood of the time delay is a simple approximation to the marginal posterior distribution of the time delay. Both Bayesian and profile likelihood approaches complement each other, producing almost identical results; the Bayesian way is more principled but the profile likelihood is easier to implement.</p>crantastic.orgtag:crantastic.org,2005:TimelineEvent/423432015-07-20T09:32:07Z2015-07-20T09:32:07Ztimedelay was released<a href="/packages/timedelay">timedelay</a> was <span class="action">released</span><br /><h3>Package description:</h3><p>We provide a toolbox to estimate the time delay between the brightness time series of gravitationally lensed quasar images via Bayesian and profile likelihood approaches. The model is based on a state-space representation for irregularly observed time series data generated from a latent continuous-time Ornstein-Uhlenbeck process. Our Bayesian method adopts scientifically motivated hyper-prior distributions and a Metropolis-Hastings within Gibbs sampler, producing posterior samples of the model parameters that include the time delay. A profile likelihood of the time delay is a simple approximation to the marginal posterior distribution of the time delay. Both Bayesian and profile likelihood approaches complement each other, producing almost identical results; the Bayesian way is more principled but the profile likelihood is easier to implement.</p>crantastic.org