Image Credit SteemWhales.com

Abstract: I have improved this NodeJs script to compute the Median and STD for the members in the Steemit-Wechat Group. It reflects the fact that: the most wealth is posseseed by a few huge whales.

I am planning to put these statistists on the Daily SteemIt Wechat Group Ranking table. The median values help better understand where you are in the big pond (are you really a minnow?)

昨天半夜写的帖子目的是小试 NodeJs 牛刀， @rivalhw 和 @dapeng 给了非常好的建议：平均值不太适合看大众水平，因为：贫富差距在 STEEM的世界里也是非常的明显，目测10%大鱼们掌握了90%的财富，所以如果取平均的话，指标都会偏高，不能反映实际情况。

是不是有很多歌唱比赛评委给分之后，选手的最终得分都得先“去掉一个最高分，去掉一个最低分”？不过在这里，你去掉一个最高最低也没有啥卵用，因为富得流油的远不止一个，穷得只剩下小裤衩的也有很多。

中位数

数学中的中位数（Median）指的是把所有需要统计的数值排序，然后取最中间那个（如果有奇数个数值），如果有偶数个，则取中间两个数来取平均。这个相对能反映出大众水平。

标准方差

标准方差 STD 反映了数据波动的情况，比如如果财富值的标准方差越大，则表示贫富差距大（离平均水平的波动区别）

NodeJs 程序 + SteemIt Wechat API

有了API，我们用 Node.Js 来跑跑，看看什么结果。首先我们通过 prototype 的方式给 Javascript 中的数组来定义以上几个指标：
Let's define these measured by extending Array's prototype in Javascript:

Array.prototype.median = function() {
    var arr = [...this];
    var len = arr.length;
    if (0 == len) {
        return NaN;
    }
    arr.sort(); 
    var mid = Math.floor(len / 2);
    if ((len % 2) == 0) {
        return (arr[mid] + arr[mid - 1]) * 0.5;
    }
    return arr[mid];
}

Array.prototype.mean = function() {
    var total = 0;
    for (var i = this.length - 1; i >= 0; -- i){
        total += this[i];
    }
    return total / this.length;
}

Array.prototype.STD = function() {
   var mean = this.mean();
   var diff = [];
   for (var j = this.length - 1; j >= 0; -- j){
       diff.push(Math.pow((this[j] - mean), 2));
   }
   return (Math.sqrt(diff.reduce((a, b) => a + b) / this.length));
}

然后一样的套路：Then we just need to fetch the ranking table via the handy API.

// @justyy
var request = require("request")
var url = "https://uploadbeta.com/api/steemit/wechat/?cached";
request({
    url: url,
    json: true
}, function (error, response, body) {
    if (!error && response.statusCode === 200) {
        var total_rep = [];
        var total_sbd = [];
        var total_steem = [];
        var total_value = [];
        var total_esp = [];
        var total_vp = [];
        var total_sp = [];
        body.forEach(function(member) {
            total_rep.push(member['rep']);
            total_sbd.push(member['sbd']);
            total_steem.push(member['steem']);
            total_value.push(member['value']);
            total_esp.push(member['esp']);
            total_vp.push(member['vp']);
            total_sp.push(member['sp']);
        });

        console.log("----Median----");
        console.log("Total Members = " + total_rep.length);
        console.log("Median Reputation = " + Math.round(total_rep.median() * 100) / 100);
        console.log("Median SBD = " + Math.round(total_sbd.median() * 100) / 100);
        console.log("Median Steem = " + Math.round(total_steem.median() * 100) / 100);
        console.log("Median Effective SP = " + Math.round(total_esp.median() * 100) / 100);
        console.log("Median SP = " + Math.round(total_sp.median() * 100) / 100);
        console.log("Median Voting Power = " + Math.round(total_vp.median() * 100) / 100);
        console.log("Median Account Value = " + Math.round(total_value.median() * 100) / 100);

        console.log("----Mean----");
        console.log("Mean Reputation = " + Math.round(total_rep.mean() * 100) / 100);
        console.log("Mean SBD = " + Math.round(total_sbd.mean() * 100) / 100);
        console.log("Mean Steem = " + Math.round(total_steem.mean() * 100) / 100);
        console.log("Mean Effective SP = " + Math.round(total_esp.mean() * 100) / 100);
        console.log("Mean SP = " + Math.round(total_sp.mean() * 100) / 100);
        console.log("Mean Voting Power = " + Math.round(total_vp.mean() * 100) / 100);
        console.log("Mean Account Value = " + Math.round(total_value.mean() * 100) / 100);      

        console.log("----STD----");
        console.log("STD Reputation = " + Math.round(total_rep.STD() * 100) / 100);
        console.log("STD SBD = " + Math.round(total_sbd.STD() * 100) / 100);
        console.log("STD Steem = " + Math.round(total_steem.STD() * 100) / 100);
        console.log("STD Effective SP = " + Math.round(total_esp.STD() * 100) / 100);
        console.log("STD SP = " + Math.round(total_sp.STD() * 100) / 100);
        console.log("STD Voting Power = " + Math.round(total_vp.STD() * 100) / 100);
        console.log("STD Account Value = " + Math.round(total_value.STD() * 100) / 100);        
    }
})

然后这是结果：And the results come out pretty quickly using NodeJs

----Median----
Total Members = 173
Median Reputation = 50.83
Median SBD = 154.62
Median Steem = 0
Median Effective SP = 28.15
Median SP = 201.16
Median Voting Power = 93.72
Median Account Value = 2040.61
----Mean----
Mean Reputation = 48.56
Mean SBD = 224.75
Mean Steem = 18.99
Mean Effective SP = 18503.56
Mean SP = 1561.65
Mean Voting Power = 83.07
Mean Account Value = 2436.73
----STD----
STD Reputation = 13.5
STD SBD = 1093.16
STD Steem = 118.9
STD Effective SP = 99019.55
STD SP = 5977.72
STD Voting Power = 21.61
STD Account Value = 9344.65

简单来说：“大众”等级是 50.83，“大众” SP 是201.16，“大众”帐号估值是 2040.61
通过STD我们可以看到，SP，ESP，还有帐号估值的贫富差距还是挺大的。

最后，再广告一下：

SteemIt 好友微信群排行榜
SteemIt 好友微信群文章列表 RSS Feed
SteemIt 编程 Geek 微信群，请联系 @justyy 让我拉你入群。

需要入群者请联系 @tumutanzi 或者 @rivalhw 谢谢。

Originally published at https://steemit.com Thank you for reading my post, feel free to Follow, Upvote, Reply, ReSteem (repost) @justyy which motivates me to create more quality posts.

原文首发于 https://Steemit.com 首发。感谢阅读，如有可能，欢迎Follow, Upvote, Reply, ReSteem (repost) @justyy 激励我创作更多更好的内容。

// Later, it will be reposted to my blogs: justyy.com, helloacm.com and codingforspeed.com 稍后同步到我的中文博客和英文计算机博客。

近期热贴

Recent Popular Posts

Tags: cn cn-programming steemit-api steemstem wechat

Sort:

Trending

[-]

tvb (65) 7 years ago

我有点不高兴的说，好多人跑来和我抢偶像......

$0.08

1 vote

justyy (74) 7 years ago

姐要不高兴了，帮你顶到最上面

$0.00

worldunderground (29) 7 years ago

Whoa, a Wechat group! Is everything in Chinese? I live in the Northeast, would love to join this, but it may be useless for me as my reading is still terrible.

tumutanzi (72) 7 years ago (edited)

As the group manager, I welcome you to join. When you chat with us in English, it is absolutely fine for me and many other steemians. We can chat with you in English. Sure, if other guys talk in Chinese, it makes no sense to you. But there is a built-in translator that may helps you.

$0.59

Yes, mainly in Chinese.

mrspointm (79) 7 years ago

辛苦YY哥啦！

braveboat (63) 7 years ago (edited)

This is pretty cool. I find it interesting, but not surprising, that many members of Steam associate by country in this crypto-netherland.

I've read a lot about WeChat when I was trading Twitter often after their IPO. Chinas definitely has many platforms that look to grow into behemoths over time and dwarf their American and European based competitors.

Upvoted and followed. Looking forward to future posts!

justyy (74) 7 years ago (edited)

Thank you. Yes, the Wechat is very big and is used by nearly 9 hundred million (if I remember correctly), so definitely this is going to be interesting if we combine wechat and steemit.

blacktranquility (64) 7 years ago

这样一来数学就全了 n_N!

其实还少了一个最基本的最小最大值。现已更新到网页中: https://helloacm.com/tools/steemit/wechat/

mumingduozi (51) 7 years ago

y哥每次都有新科技，支持中位数的想法！

哈哈，还好不是 “黑科技”

kenchung (66) 7 years ago

這裡太多神人了...

太多大鱼了，STEEM人才济济。

rivalhw (80) 7 years ago

这次的数据就很有参考意义了，也比较能真实地反映出目前大多数民众的状况

多谢大伟哥的肯定和建议！

tumutanzi (72) 7 years ago

计算机专家变身数学家。

其实我数学并不怎么样~ 不过学计算机的多少数学逻辑还凑合

minnowpond1 (48) 7 years ago

This post recieved an upvote from minnowpond. If you would like to recieve upvotes from minnowpond on all your posts, simply FOLLOW @minnowpond

The Average, Median, STD of SteemIt Wechat Group 数据初步分析系列 STEEM中文微信群排行榜单 - 中位数，平均，和标准方差

中位数

标准方差

NodeJs 程序 + SteemIt Wechat API

近期热贴

Recent Popular Posts