Description Usage Arguments Value Author(s) See Also Examples

This function estimates maximum likelihood models (e.g., Poisson or Logit) and is efficient to handle fixed effects (i.e. cluster variables). It further allows for nonlinear right hand sides.

1 2 3 4 |

`linear.fml` |
A formula. The linear formula to be estimated. |

`family` |
Character scalar. It should provide the family. Currently |

`data` |
A data.frame containing the necessary variables to run the model. The variables of the non-linear right hand side of the formula are identified with this data.frame names. Note that no NA is allowed. |

`start` |
A list. Starting values for the non-linear parameters. ALL the parameters are to be named and given a staring value. Example: |

`dummy` |
Character vector. The name/s of a/some variable/s within the dataset. These variables should contain the identifier of each observation (e.g., think of it as a panel identifier). |

`linear.start` |
Numeric named vector. The starting values of the linear part. If it is just a numeric scalar, all coefficients are set to |

`useHessian` |
Logical. (Only if omptimization method is |

`opt_method` |
Character scalar. Which optimization method should be used. Either |

`opt.control` |
List of elements to be passed to the optimization method (nlminb or optim). |

`optim.method` |
Character scalar. If |

`debug` |
Logical. If |

`...` |
Not currently used. |

An `feNmlm`

object.

`coef` |
The coefficients. |

`coeftable` |
The table of the coefficients with their standard errors, z-values and p-values. |

`loglik` |
The loglikelihood. |

`iterations` |
Number of iterations of the algorithm. |

`n` |
The number of observations. |

`k` |
The number of parameters of the model. |

`call` |
The call. |

`nonlinear.fml` |
The nonlinear formula of the call. It also contains the dependent variable. |

`linear.formula` |
The linear formula of the call. |

`ll_null` |
Log-likelyhood of the null model |

`pseudo_r2` |
The adjusted pseudo R2. |

`naive.r2` |
The R2 as if the expected predictor was the linear predictor in OLS. |

`message` |
The convergence message from the optimization procedures. |

`sq.cor` |
Squared correlation between the dependent variable and its expected value as given by the optimization. |

`expected.predictor` |
The expected predictor is the expected value of the dependent variable. |

`cov.unscaled` |
The variance covariance matrix of the parameters. |

`sd` |
The standard error of the parameters. |

Laurent Berge

See also `feNmlm`

.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 | ```
#The data
n = 100
x = rnorm(n,1,5)**2
y = rnorm(n,-1,5)**2
z = rpois(n,x*y)
base = data.frame(x,y,z)
# Results of the Poisson..
est_poisson = femlm(z~log(x)+log(y),base,family="poisson")
# .. and of the Negative Binomial
est_negbin = femlm(z~log(x)+log(y),base,family="negbin")
# Displaying the results
est_poisson
est_negbin
# Changing the way the standard errors are computed:
summary(est_poisson,sd="white")
summary(est_negbin,sd="white")
#
# Now with dummies
#
# Bilateral network
nb = 20
n = nb**2
k = nb
id1 = factor(rep(1:k,each=n/k))
id2 = factor(rep(1:(n/k),times=k))
x = rnorm(n,1,5)**2
y = rnorm(n,-1,5)**2
z = rpois(n,x*y+rnorm(n,sd = 3)**2)
base = data.frame(x,y,z,id1,id2)
# We want to use the ID's of each observation as a variable: we use the option dummy
est_poisson = femlm(z~log(x)+log(y),base,family="poisson",dummy=c("id1","id2"))
# Displaying the results with twoway clustered santard-errors
print(est_poisson,"t")
``` |

FENmlm documentation built on May 29, 2017, 2:46 p.m.

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.