6.10
Distribuzione normale
La distribuzione normale, ovvero continua, è la più importante di tutte le distribuzioni. Il suo grafico è una curva simmetrica a forma di campana, ch…
La distribuzione normale è una distribuzione di probabilità continua con un grafico simmetrico a forma di campana. È descritto dalla formula di distribuzione gaussiana con media e deviazione standard come parametri fissi. Le π ed e sono valori costanti.
Considera il peso alla nascita dei bambini con una media di 3,5 kg e una deviazione standard di 0,4 kg. I dati possono essere visualizzati tracciando la densità di probabilità rispetto al peso alla nascita.
Dalla formula del punteggio z, i pesi alla nascita possono essere standardizzati nei corrispondenti punteggi z.
La ridefinizione della densità di probabilità con il punteggio z mostra che il grafico è ora centrato intorno allo zero.
Questa forma standardizzata della distribuzione normale è nota come distribuzione normale standard, in cui la media è zero e la deviazione standard è uno.
Tale conversione della distribuzione normale in distribuzione normale standard semplifica la formula della distribuzione gaussiana, facilitando il calcolo dei valori di probabilità.
È anche utile per confrontare set di dati con medie e deviazioni standard diverse.
View the full transcript and gain access to JoVE Core videos
Q1: What is a normal distribution and why is it important?
The normal distribution is a continuous probability distribution with a symmetrical, bell-shaped graph described by the Gaussian distribution formula. It is the most important distribution across disciplines including psychology, business, economics, sciences, and nursing. The curve is symmetric about the mean, and the area under the curve equals one. Most real-world phenomena like IQ scores and real-estate prices follow this distribution.
Q2: How do mean and standard deviation affect the shape of a normal distribution?
The normal distribution depends only on two parameters: mean (μ) and standard deviation (σ). Changes in standard deviation alter the curve's shape, making it fatter or skinnier. Changes in mean shift the graph left or right. This means there are infinite normal probability distributions, each uniquely defined by its mean and standard deviation values.
Q3: What is the standard normal distribution and how is it created?
The standard normal distribution is a normal distribution of standardized values called z scores, where the mean is zero and standard deviation is one. Converting a normal distribution into standard normal distribution simplifies the Gaussian distribution formula and eases probability calculations. This standardization also allows comparison of datasets with different means and standard deviations.
Q4: How do you calculate a z score from a normal distribution?
A z score is measured in units of standard deviation from the mean. For example, if the mean is five and standard deviation is two, the value 11 is three standard deviations above the mean, giving a z score of 3. The z score formula standardizes any value from a normal distribution, enabling comparison across different datasets and probability calculations.
Q5: What does symmetry mean in a normal distribution?
A normal distribution is symmetric about a vertical line drawn through the mean. This symmetry means the mean equals the median in theory, as the curve mirrors itself on both sides. The bell-shaped graph reflects this perfect balance, with equal proportions of data distributed on either side of the center value.
Q6: Can the normal distribution be applied to all real-world data?
Although the normal distribution is extremely important and widely observed, it cannot be applied to everything in the real world. While many phenomena like birth weights, test scores, and prices follow this pattern, not all datasets conform to a bell-shaped curve. Understanding when the normal distribution applies is essential for accurate statistical analysis and interpretation.
Q7: Why is standardizing data into z scores useful?
Standardizing data into z scores simplifies calculations and enables meaningful comparisons between datasets with different means and standard deviations. The z score formula converts any normal distribution value into the standard normal distribution framework, where the mean is zero and standard deviation is one. This uniform scale makes probability calculations and data interpretation more straightforward across diverse applications.