Trimmed and Winsorized Means

/**
 * Fintech Math Bootcamp · Lesson 035 of 120
 * Trimmed and Winsorized Means
 * Module 04: Location, Ranking, and Exploratory Summaries
 *
 * Scenario: Reducing outlier influence without pretending data vanished
 * Rule:     trim drops tails; winsorization caps tails
 *
 * Try it:   Which operation preserves the number of observations?
 *
 * Lesson article: https://thefintechbuilder.com/financial-mathematics-statistics-and-data-foundations/location-ranking-and-exploratory-summaries/trimmed-and-winsorized-means/
 * Free course:    https://courses.thefintechbuilder.com
 * Synthetic teaching example, not financial advice or a production library.
 */

export function lesson035() {
  const x = [1, 2, 2, 4, 9].sort((a,b) => a-b);
  const k = Math.floor(x.length * 0.2); // 20% from each tail
  const trimmed = x.slice(k, x.length - k);
  const lo = x[k], hi = x[x.length - 1 - k];
  const winsorized = x.map(v => Math.max(lo, Math.min(hi, v)));
  const avg = (a: number[]) => a.reduce((s,v)=>s+v,0)/a.length;
  const result = {trimmedMean: avg(trimmed),
    winsorizedMean: avg(winsorized)};
  return result;
}

export const checkedResult = {"trimmedMean":2.6666666666666665,"winsorizedMean":2.8};

// Run this file directly: npx tsx lessons/04-location-ranking-and-exploratory-summaries/035-trimmed-and-winsorized-means.ts
if (process.argv[1] && import.meta.url.endsWith(process.argv[1].replace(/\\/g, "/").split("/").pop()!)) {
  console.log(JSON.stringify(lesson035(), null, 2));
}

# Fintech Math Bootcamp · Lesson 035 of 120
# Trimmed and Winsorized Means
# Module 04: Location, Ranking, and Exploratory Summaries
#
# Scenario: Reducing outlier influence without pretending data vanished
# Rule:     trim drops tails; winsorization caps tails
#
# Try it:   Which operation preserves the number of observations?
#
# Lesson article: https://thefintechbuilder.com/financial-mathematics-statistics-and-data-foundations/location-ranking-and-exploratory-summaries/trimmed-and-winsorized-means/
# Free course:    https://courses.thefintechbuilder.com
# Synthetic teaching example, not financial advice or a production library.

import json
import math


def lesson035() -> dict:
    x = sorted([1, 2, 2, 4, 9])
    k = math.floor(len(x) * 0.2)  # 20% from each tail
    trimmed = x[k:len(x) - k]
    lo, hi = x[k], x[len(x) - 1 - k]
    winsorized = [max(lo, min(hi, v)) for v in x]

    def avg(a: list) -> float:
        return sum(a) / len(a)

    return {"trimmedMean": avg(trimmed), "winsorizedMean": avg(winsorized)}


if __name__ == "__main__":
    print(json.dumps(lesson035(), indent=2))

/**
 * Fintech Math Bootcamp · Lesson 035 of 120
 * Trimmed and Winsorized Means
 * Module 04: Location, Ranking, and Exploratory Summaries
 *
 * Scenario: Reducing outlier influence without pretending data vanished
 * Rule:     trim drops tails; winsorization caps tails
 *
 * Try it:   Which operation preserves the number of observations?
 *
 * Lesson article: https://thefintechbuilder.com/financial-mathematics-statistics-and-data-foundations/location-ranking-and-exploratory-summaries/trimmed-and-winsorized-means/
 * Free course:    https://courses.thefintechbuilder.com
 * Synthetic teaching example, not financial advice or a production library.
 */

import java.util.ArrayList;
import java.util.Arrays;
import java.util.LinkedHashMap;
import java.util.List;
import java.util.Map;

public class Main {
    static double avg(double[] a) {
        double s = 0;
        for (double v : a) s += v;
        return s / a.length;
    }

    static Map<String, Object> lesson035() {
        double[] x = {1, 2, 2, 4, 9};
        Arrays.sort(x);
        int k = (int) Math.floor(x.length * 0.2); // 20% from each tail
        double[] trimmed = Arrays.copyOfRange(x, k, x.length - k);
        double lo = x[k], hi = x[x.length - 1 - k];
        double[] winsorized = new double[x.length];
        for (int i = 0; i < x.length; i++) winsorized[i] = Math.max(lo, Math.min(hi, x[i]));

        Map<String, Object> result = new LinkedHashMap<String, Object>();
        result.put("trimmedMean", avg(trimmed));
        result.put("winsorizedMean", avg(winsorized));
        return result;
    }

    public static void main(String[] args) {
        System.out.println(toJson(lesson035(), ""));
    }

    // --- Minimal JSON printer: maps keep insertion order, 2-space indent. ---

    static String toJson(Object value, String indent) {
        if (value == null) return "null";
        if (value instanceof String) return "\"" + value + "\"";
        if (value instanceof Double) return formatNumber((Double) value);
        if (value instanceof Number) return value.toString();
        if (value instanceof double[]) {
            List<Object> items = new ArrayList<Object>();
            for (double v : (double[]) value) items.add(v);
            return toJson(items, indent);
        }
        String inner = indent + "  ";
        StringBuilder sb = new StringBuilder();
        if (value instanceof Map) {
            Map<?, ?> map = (Map<?, ?>) value;
            if (map.isEmpty()) return "{}";
            sb.append("{\n");
            int i = 0;
            for (Map.Entry<?, ?> entry : map.entrySet()) {
                sb.append(inner).append('"').append(entry.getKey()).append("\": ")
                  .append(toJson(entry.getValue(), inner))
                  .append(++i < map.size() ? ",\n" : "\n");
            }
            return sb.append(indent).append('}').toString();
        }
        List<?> list = (List<?>) value;
        if (list.isEmpty()) return "[]";
        sb.append("[\n");
        for (int i = 0; i < list.size(); i++) {
            sb.append(inner).append(toJson(list.get(i), inner))
              .append(i + 1 < list.size() ? ",\n" : "\n");
        }
        return sb.append(indent).append(']').toString();
    }

    static String formatNumber(double v) {
        if (Double.isNaN(v) || Double.isInfinite(v)) return "null";
        if (v == Math.rint(v) && Math.abs(v) < 1e15) return Long.toString((long) v);
        return Double.toString(v);
    }
}

// Fintech Math Bootcamp · Lesson 035 of 120
// Trimmed and Winsorized Means
// Module 04: Location, Ranking, and Exploratory Summaries
//
// Scenario: Reducing outlier influence without pretending data vanished
// Rule:     trim drops tails; winsorization caps tails
//
// Try it:   Which operation preserves the number of observations?
//
// Lesson article: https://thefintechbuilder.com/financial-mathematics-statistics-and-data-foundations/location-ranking-and-exploratory-summaries/trimmed-and-winsorized-means/
// Free course:    https://courses.thefintechbuilder.com
// Synthetic teaching example, not financial advice or a production library.

package main

import (
	"encoding/json"
	"fmt"
	"math"
	"sort"
)

// RobustMeans compares dropping the tails with capping them.
type RobustMeans struct {
	TrimmedMean    float64 `json:"trimmedMean"`
	WinsorizedMean float64 `json:"winsorizedMean"`
}

func avg(a []float64) float64 {
	s := 0.0
	for _, v := range a {
		s += v
	}
	return s / float64(len(a))
}

func lesson035() RobustMeans {
	x := []float64{1, 2, 2, 4, 9}
	sort.Float64s(x)
	k := int(math.Floor(float64(len(x)) * 0.2)) // 20% from each tail
	trimmed := x[k : len(x)-k]
	lo, hi := x[k], x[len(x)-1-k]
	winsorized := make([]float64, len(x))
	for i, v := range x {
		winsorized[i] = math.Max(lo, math.Min(hi, v))
	}
	return RobustMeans{TrimmedMean: avg(trimmed), WinsorizedMean: avg(winsorized)}
}

func main() {
	out, _ := json.MarshalIndent(lesson035(), "", "  ")
	fmt.Println(string(out))
}

/**
 * Fintech Math Bootcamp · Lesson 035 of 120
 * Trimmed and Winsorized Means
 * Module 04: Location, Ranking, and Exploratory Summaries
 *
 * Scenario: Reducing outlier influence without pretending data vanished
 * Rule:     trim drops tails; winsorization caps tails
 *
 * Try it:   Which operation preserves the number of observations?
 *
 * Lesson article: https://thefintechbuilder.com/financial-mathematics-statistics-and-data-foundations/location-ranking-and-exploratory-summaries/trimmed-and-winsorized-means/
 * Free course:    https://courses.thefintechbuilder.com
 * Synthetic teaching example, not financial advice or a production library.
 */

#include <algorithm>
#include <charconv>
#include <cmath>
#include <iostream>
#include <numeric>
#include <string>
#include <utility>
#include <vector>

// --- Minimal JSON value and printer: objects keep insertion order, 2-space indent. ---

struct Json {
    enum class Kind { Null, Number, Text, Array, Object };
    Kind kind = Kind::Null;
    double number = 0;
    std::string text;
    std::vector<std::string> keys;  // object keys, parallel to items
    std::vector<Json> items;        // array elements or object values
};

Json num(double v) { Json j; j.kind = Json::Kind::Number; j.number = v; return j; }
Json str(const std::string& s) { Json j; j.kind = Json::Kind::Text; j.text = s; return j; }
Json arr(const std::vector<Json>& values) { Json j; j.kind = Json::Kind::Array; j.items = values; return j; }

Json arr(const std::vector<double>& values) {
    std::vector<Json> items;
    for (double v : values) items.push_back(num(v));
    return arr(items);
}

Json obj(const std::vector<std::pair<std::string, Json>>& fields) {
    Json j;
    j.kind = Json::Kind::Object;
    for (const auto& [key, value] : fields) { j.keys.push_back(key); j.items.push_back(value); }
    return j;
}

std::string formatNumber(double v) {
    if (!std::isfinite(v)) return "null";
    char buf[64];
    auto end = std::to_chars(buf, buf + sizeof buf, v).ptr;  // shortest round-trip form
    return std::string(buf, end);
}

void writeJson(std::ostream& out, const Json& j, const std::string& indent) {
    switch (j.kind) {
    case Json::Kind::Null: out << "null"; return;
    case Json::Kind::Number: out << formatNumber(j.number); return;
    case Json::Kind::Text: out << '"' << j.text << '"'; return;
    default: break;
    }
    bool isObject = j.kind == Json::Kind::Object;
    if (j.items.empty()) { out << (isObject ? "{}" : "[]"); return; }
    std::string inner = indent + "  ";
    out << (isObject ? "{\n" : "[\n");
    for (size_t i = 0; i < j.items.size(); ++i) {
        out << inner;
        if (isObject) out << '"' << j.keys[i] << "\": ";
        writeJson(out, j.items[i], inner);
        out << (i + 1 < j.items.size() ? ",\n" : "\n");
    }
    out << indent << (isObject ? '}' : ']');
}

// --- Lesson ---

double avg(const std::vector<double>& a) {
    return std::accumulate(a.begin(), a.end(), 0.0) / a.size();
}

Json lesson035() {
    std::vector<double> x = {1, 2, 2, 4, 9};
    std::sort(x.begin(), x.end());
    size_t k = static_cast<size_t>(std::floor(x.size() * 0.2));  // 20% from each tail
    std::vector<double> trimmed(x.begin() + k, x.end() - k);
    double lo = x[k], hi = x[x.size() - 1 - k];
    std::vector<double> winsorized;
    for (double v : x) winsorized.push_back(std::max(lo, std::min(hi, v)));
    return obj({{"trimmedMean", num(avg(trimmed))}, {"winsorizedMean", num(avg(winsorized))}});
}

int main() {
    writeJson(std::cout, lesson035(), "");
    std::cout << '\n';
}

// Fintech Math Bootcamp · Lesson 035 of 120
// Trimmed and Winsorized Means
// Module 04: Location, Ranking, and Exploratory Summaries
//
// Scenario: Reducing outlier influence without pretending data vanished
// Rule:     trim drops tails; winsorization caps tails
//
// Try it:   Which operation preserves the number of observations?
//
// Lesson article: https://thefintechbuilder.com/financial-mathematics-statistics-and-data-foundations/location-ranking-and-exploratory-summaries/trimmed-and-winsorized-means/
// Free course:    https://courses.thefintechbuilder.com
// Synthetic teaching example, not financial advice or a production library.

// --- Minimal JSON value and printer: objects keep insertion order, 2-space indent. ---

#[allow(dead_code)]
enum Json {
    Null,
    Num(f64),
    Str(String),
    Arr(Vec<Json>),
    Obj(Vec<(String, Json)>),
}

#[allow(dead_code)]
fn nums(values: &[f64]) -> Json {
    Json::Arr(values.iter().map(|&v| Json::Num(v)).collect())
}

#[allow(dead_code)]
fn obj(fields: Vec<(&str, Json)>) -> Json {
    Json::Obj(fields.into_iter().map(|(k, v)| (k.to_string(), v)).collect())
}

fn format_number(v: f64) -> String {
    if !v.is_finite() {
        return "null".to_string();
    }
    if v.fract() == 0.0 && v.abs() < 1e15 {
        return format!("{}", v as i64);
    }
    format!("{:?}", v) // shortest round-trip form
}

impl Json {
    fn render(&self, indent: &str) -> String {
        let inner = format!("{}  ", indent);
        match self {
            Json::Null => "null".to_string(),
            Json::Num(v) => format_number(*v),
            Json::Str(s) => format!("\"{}\"", s),
            Json::Arr(items) if items.is_empty() => "[]".to_string(),
            Json::Obj(fields) if fields.is_empty() => "{}".to_string(),
            Json::Arr(items) => {
                let lines: Vec<String> = items.iter().map(|v| format!("{}{}", inner, v.render(&inner))).collect();
                format!("[\n{}\n{}]", lines.join(",\n"), indent)
            }
            Json::Obj(fields) => {
                let lines: Vec<String> = fields
                    .iter()
                    .map(|(k, v)| format!("{}\"{}\": {}", inner, k, v.render(&inner)))
                    .collect();
                format!("{{\n{}\n{}}}", lines.join(",\n"), indent)
            }
        }
    }
}

// --- Lesson ---

fn avg(a: &[f64]) -> f64 {
    a.iter().sum::<f64>() / a.len() as f64
}

fn lesson035() -> Json {
    let mut x: Vec<f64> = vec![1.0, 2.0, 2.0, 4.0, 9.0];
    x.sort_by(|a, b| a.partial_cmp(b).unwrap());
    let k = (x.len() as f64 * 0.2).floor() as usize; // 20% from each tail
    let trimmed = &x[k..x.len() - k];
    let (lo, hi) = (x[k], x[x.len() - 1 - k]);
    let winsorized: Vec<f64> = x.iter().map(|&v| lo.max(hi.min(v))).collect();
    obj(vec![
        ("trimmedMean", Json::Num(avg(trimmed))),
        ("winsorizedMean", Json::Num(avg(&winsorized))),
    ])
}

fn main() {
    println!("{}", lesson035().render(""));
}

/**
 * Fintech Math Bootcamp · Lesson 035 of 120
 * Trimmed and Winsorized Means
 * Module 04: Location, Ranking, and Exploratory Summaries
 *
 * Scenario: Reducing outlier influence without pretending data vanished
 * Rule:     trim drops tails; winsorization caps tails
 *
 * Try it:   Which operation preserves the number of observations?
 *
 * Lesson article: https://thefintechbuilder.com/financial-mathematics-statistics-and-data-foundations/location-ranking-and-exploratory-summaries/trimmed-and-winsorized-means/
 * Free course:    https://courses.thefintechbuilder.com
 * Synthetic teaching example, not financial advice or a production library.
 */

using System;
using System.Linq;
using System.Text.Json;

var options = new JsonSerializerOptions { WriteIndented = true };
Console.WriteLine(JsonSerializer.Serialize(Lesson035(), options));

static object Lesson035()
{
    double[] x = new double[] { 1, 2, 2, 4, 9 }.OrderBy(v => v).ToArray();
    int k = (int)Math.Floor(x.Length * 0.2); // 20% from each tail
    double[] trimmed = x[k..(x.Length - k)];
    double lo = x[k], hi = x[x.Length - 1 - k];
    double[] winsorized = x.Select(v => Math.Max(lo, Math.Min(hi, v))).ToArray();
    return new { trimmedMean = trimmed.Average(), winsorizedMean = winsorized.Average() };
}