approximates a(x) using interval midpoints (except a0, which is estimated using a rule-of-thumb formula scaled by m0)

1 | ```
axMidpoint(Mx, n)
``` |

`Mx` |
a numeric vector of central death rates, defined as deaths/exposure. |

`n` |
a numeric vector of the age-interval widths. Must be the same length as Mx. |

All values will be in the exact middle of the age interval, except for a0, which uses: `a0 = .07 + 1.7 * M0`

.

returns a numeric vector of ax values

Tim Riffe

formula for a0 as suggested by: Keyfitz, N. (1970), Finding probabilities from observed rates, or how to make a life table. The American Statistician (1970), pp. 28-33

See Also `axEstimate`

, a wrapper function for this and 3 other a(x) estimation methods (`axKeyfitz`

,`axSchoen`

and `axPreston`

).

Questions? Problems? Suggestions? Tweet to @rdrrHQ or email at ian@mutexlabs.com.

All documentation is copyright its authors; we didn't write any of that.