Statistician, map-maker, dataviz consultant and trainer, Observable ambassador & R instructor. I enjoy explaining things i love in detail, the most clearly i can, and building live analytical opendataflows.
@ericmauviere
x_axis = [
Plot.ruleY([0], { x1: 0, x2: 800, dx: gap, strokeWidth: 0.5, stroke:'#999' }),
Plot.ruleX(ticks, { x: d => d, y: 0, insetBottom: -5, dx: gap, stroke:'#999' }),
Plot.text(ticks, { x: d => d, y: 0, dx: gap, dy: 12 }),
Plot.ruleY([0], { x1: 0, x2: -800, dx: -gap, strokeWidth: 0.5, stroke:'#999' }),
Plot.ruleX(ticks, { x: d => -d, y: 0, insetBottom: -5, dx: -gap, stroke:'#999' }),
Plot.text(ticks, { x: d => -d, y: 0, dx: -gap, dy: 12 }),
Plot.text(['← Männer in Tausend'], { x: 0, y: 0, dx: -gap, dy: 25, textAnchor: 'end', fontWeight: 'bold' }),
Plot.text(['Frauen in Tausend → '], { x: 0, y: 0, dx: gap, dy: 25, textAnchor: 'start', fontWeight: 'bold' })
]
y_axis = [
Plot.text(d3.range(0, 90, 10), { y: d => d }),
Plot.text(["Jahre"], { y: 80, dy: 10, fontWeight: 400 }),
Plot.ruleY(d3.range(5, 85, 5), { x1: d => d % 10 == 0 ? 0 : gap, x2: 100, dx: gap/2, strokeWidth: 0.2 }),
Plot.ruleY(d3.range(5, 85, 5), { x1: d => d % 10 == 0 ? 0 : -gap, x2: -100, dx: -gap/2, strokeWidth: 0.2 })
]
data_pivot = aq.from(data_long)
.derive({ Zensus: "'y_' + d.Zensus" })
.groupby("Alter_in_Jahren", "sexe", "categ")
.pivot("Zensus", "value")
interpol = (l) => {
let t = 2 * l - 1
return data_pivot.derive({
value: `Math.max(0, ${(t * (t - 1)) / 2} * d.y_1991
+ ${-(t + 1) * (t - 1)} * d.y_2006
+ ${((t + 1) * t) / 2} * d.y_2021)`
}) // Lagrange (quadratic) interpolation
}
interpol(0).orderby(0,1).view()
mortality_rate_by_age_sex = (age, sexe) => { // Germany mortality table, 2005: https://apps.who.int/gho/data/view.main.LT62050
if (age < 40) return 1
if (age < 45) return sexe == "M" ? 2 : 1
if (age < 50) return sexe == "M" ? 2.6 : 1.4
if (age < 55) return sexe == "M" ? 6 : 3
if (age < 60) return sexe == "M" ? 8 : 4
if (age < 65) return sexe == "M" ? 13 : 6
if (age < 70) return sexe == "M" ? 20 : 10
if (age < 75) return sexe == "M" ? 32 : 17
if (age < 80) return sexe == "M" ? 54 : 32
if (age < 85) return sexe == "M" ? 90 : 62
}
interpol3 = l => { // aging pyramids
let t = 2 * l
aq.addFunction('decrease_with_aging', decrease_with_aging, { override: true })
aq.addFunction('increase_with_getting_younger', increase_with_getting_younger, { override: true })
if (l <= 0.5) {
let decal = Math.round(15 * t), decal2 = 15 - decal
let dt1991 = data_pivot.select(0,1,2,'y_1991')
.derive({ y_1991: `op.decrease_with_aging(d.Alter_in_Jahren, d.sexe, ${decal}, d.y_1991)`} )
.derive({ Alter_in_Jahren: `d.Alter_in_Jahren + ${decal}` } )
.rename({ 'y_1991': 'y_1991_d' })
let dt1996 = data_pivot.select(0,1,2,'y_2006')
.derive({ Alter_in_Jahren: `d.Alter_in_Jahren - ${decal2}`} )
.derive({ y_2006: `op.increase_with_getting_younger(d.Alter_in_Jahren, d.sexe, ${decal2}, d.y_2006)`} )
.rename({ 'y_2006': 'y_2006_d' })
return dt1991.join_full(dt1996)
.orderby('Alter_in_Jahren').filter('d.Alter_in_Jahren >= 0 && d.Alter_in_Jahren < 85')
.derive({ value: `d.y_1991_d === undefined ? d.y_2006_d :
d.y_2006_d === undefined ? d.y_1991_d :
((1 - ${t*t}) * d.y_1991_d + ${t*t} * d.y_2006_d)`})
} else {
t = 2 * (l - 0.5)
let decal = Math.round(15 * t), decal2 = 15 - decal
let dt2006 = data_pivot.select(0,1,2,'y_2006')
.derive({ y_2006: `op.decrease_with_aging(d.Alter_in_Jahren, d.sexe, ${decal}, d.y_2006)`} )
.derive({ Alter_in_Jahren: `d.Alter_in_Jahren + ${decal}`})
.rename({ 'y_2006': 'y_2006_d' })
let dt2021 = data_pivot.select(0,1,2,'y_2021')
.derive({ Alter_in_Jahren: `d.Alter_in_Jahren - ${decal2}`} )
.derive({ y_2021: `op.increase_with_getting_younger(d.Alter_in_Jahren, d.sexe, ${decal2}, d.y_2021)`} )
.rename({ 'y_2021': 'y_2021_d' })
return dt2006.join_full(dt2021)
.orderby('Alter_in_Jahren').filter('d.Alter_in_Jahren >= 0 && d.Alter_in_Jahren < 85')
.derive({ value: `d.y_2006_d === undefined ? d.y_2021_d :
d.y_2021_d === undefined ? d.y_2006_d :
((1 - ${t*t}) * d.y_2006_d + ${t*t} * d.y_2021_d)`})
}
}
interpol3(0).view()
//https://www.ined.fr/fr/tout-savoir-population/chiffres/france/mortalite-cause-deces/taux-mortalite-sexe-age/
Purpose-built for displays of data
Observable is your go-to platform for exploring data and creating expressive data visualizations. Use reactive JavaScript notebooks for prototyping and a collaborative canvas for visual data exploration and dashboard creation.
