Description Usage Arguments Value See Also Examples

`susie_auto`

is an attempt to automate reliable
running of susie even on hard problems. It implements a three-stage
strategy for each L: first, fit susie with very small residual
error; next, estimate residual error; finally, estimate the prior
variance. If the last step estimates some prior variances to be
zero, stop. Otherwise, double L, and repeat. Initial runs are
performed with relaxed tolerance; the final run is performed using
the default susie tolerance.

1 2 3 4 5 6 7 8 9 10 11 12 13 | ```
susie_auto(
X,
y,
L_init = 1,
L_max = 512,
verbose = FALSE,
init_tol = 1,
standardize = TRUE,
intercept = TRUE,
max_iter = 100,
tol = 0.01,
...
)
``` |

`X` |
An n by p matrix of covariates. |

`y` |
The observed responses, a vector of length n. |

`L_init` |
The initial value of L. |

`L_max` |
The largest value of L to consider. |

`verbose` |
If |

`init_tol` |
The tolerance to passed to |

`standardize` |
If |

`intercept` |
If |

`max_iter` |
Maximum number of IBSS iterations to perform. |

`tol` |
A small, non-negative number specifying the convergence
tolerance for the IBSS fitting procedure. The fitting procedure
will halt when the difference in the variational lower bound, or
“ELBO” (the objective function to be maximized), is
less than |

`...` |
Additional arguments passed to |

See `susie`

for a description of return values.

1 2 3 4 5 6 7 8 9 10 11 12 13 | ```
set.seed(1)
n = 1000
p = 1000
beta = rep(0,p)
beta[1:4] = 1
X = matrix(rnorm(n*p),nrow = n,ncol = p)
X = scale(X,center = TRUE,scale = TRUE)
y = drop(X %*% beta + rnorm(n))
res = susie_auto(X,y)
plot(beta,coef(res)[-1])
abline(a = 0,b = 1,col = "skyblue",lty = "dashed")
plot(y,predict(res))
abline(a = 0,b = 1,col = "skyblue",lty = "dashed")
``` |

Embedding an R snippet on your website

Add the following code to your website.

For more information on customizing the embed code, read Embedding Snippets.