shijun0210

一家 2020 年才起步的广州家具品牌，用「全装配交付」这一个反行业惯例的动作，在美国中高端家具市场撕开缺口——客单价 1,800 美元以上、2025 年线上销售约 7,000 万美元，并在洛杉矶西好莱坞开出全球首家旗舰店。 本文拆解它的背景、行业处境、商业模型与竞争格局。 执行摘要先用六个数字建立坐标： 指标 数值 说明 2025 线上销售额 约 7,000 万美元 约 6 亿元人民币，成立第 5 年 平均客单价 1,800 美元+ 亚马逊等渠道约 2,000 美元，高于 95% 的家具出海品牌 上新周期 45 天 对比 West Elm / CB2 一年仅两季、约 30 个 SKU 北美履约时效 48 小时出库 覆盖 90% 订单，自建仓配体系 质保期 5 年 北美同业普遍为 1 年 营收量级 约 0.7 亿美元 处于跨境家居品牌的腰部 五个核心判断1 · 它卖的不是家具，是「省掉的那 50 分钟」 北美家具半个世纪的交付标准是宜家定义的平板包装＋自行组装（45–60 分钟）。POVISON...

【2026-1期】 Digital Resilience for the Climate Crisis: A Multi-Perspective Analysishttps://doi.org/10.25300/MISQ/2025/18779 This commentary explores multiple perspectives on the potential use of digital technologies to improve organizational resilience in the context of climate change. Such an approach is needed to address this complex problem space, especially since it encompasses a wide variety of phenomena, including floods and landslides, disruptions to global supply chains, heat waves,...

【2026-2期】 Regulating AI: Lessons From Scientific Computinghttps://doi.org/10.25300/MISQ/2025/19111 Motivated by the perceived importance of explainability to AI regulation, this paper examines the complex relationship between theory, as a proxy for explainability, and computation in scientific practice. Drawing from both historical and contemporary examples of scientific computing, we make three key arguments: First, we show that theory-computation relationships exist in diverse...

数据分类按数据对象分类参考数据参考数据是指对其他数据进行分类和规范的数据，如国家、地区、货币、计量单位等产业通用的数据及各产业特色基础配置数据。为了简化，有的企业称这类数据为配置型主数据，也有的企业称这类数据为通用基础类数据。它是相对稳定、静态的数据，基本上不会变化，往往通过系统配置文件给予规范并固化在信息管理系统中。 主数据主数据是指满足跨部门业务协同需要的、反映核心业务实体状态属性的基础信息。主数据是用来描述企业核心业务实体的数据，是企业核心业务对象、交易业务的执行主体，是在整个价值链上被重复或共享应用于多个业务流程、跨越多个业务部门和系统、高价值的基础数据，也是各业务应用和各系统之间进行数据交互的基础。从业务角度看，主数据是相对固定、变化缓慢的，但它是企业信息系统的神经中枢，是业务运行和决策分析的基础，如供应商、客户、企业组织机构和员工、产品、客户、供应商、物料等主数据 ...

描述性统计用途：数据分布探索、异常值检测 12summary(data) # 五数概括psych::describe(data) # 详细统计量（偏度/峰度） 假设检验T检验（两组均值差异）： 1t.test(group1, group2, var.equal=TRUE) # 独立样本 卡方检验（分类变量关联性）： 1chisq.test(table(var1, var2)) 回归模型 [!NOTE] 注意 模型前提检验：回归需验证正态性、异方差性；PCA/FA需KMO>0.6 可视化优先：用 ggplot2/plotly 探索数据再建模 可重复性：结合 R Markdown 或 Quarto 生成动态报告 线性回归（LM）123model <- lm(y ~ x1 + x2, data=df)summary(model) # 系数显著性car::vif(model) # 多重共线性检验 广义线性模型（GLM）12345# Logistic回归（二分类因变量）glm(y_binary ~ x1...

大模型微调，指的是在已有的大规模预训练模型基础上，通过对标注数据进行训练，进一步优化模型的表现，以适应特定任务或场景的需求。 RAG/Agent技术：通过搭建工作流来优化模型表现微调：通过修改模型参数来优化模型表现 RAG Retrieval-Augmented Generation 检索-增强-生成 分片：将文档切片成多个片段 索引：通过Embedding将片段文本转换为向量，然后将片段文本和片段向量存入向量数据库中 Embedding模型排行榜：mteb，目前top5（截至25.7.28）：gemini-embedding-001Qwen3-Embedding-8BQwen3-Embedding-4BQwen3-Embedding-0.6BLinq-Embed-Mistral 召回（初步筛选）：从片段里面挑出与用户输入最相似的n个部分。计算向量相似度，常用计算方法：余弦相似度、欧氏距离、点积 重排（精挑细选）：从召回的n个部分，再挑出k个最相似的。计算方法...

Many Models of Inequality（长尾-帕累托分布-幂律分布）We start by summarizing some empirical regularities. First, in all countries at all times, the distribution of income has an elongated tail, with many low-income people and a small percentage of people who earn large incomes. Historically,income distributions were calibrated to lognormal distribution or Pareto distributions. Recently, more granular data reveals the tail to be longer than lognormal, though not quite that described by a power law. Wealth...

Multi-Armed Bandit Problems 多臂老虎机问题（不确定+最优备选）we add uncertainty to the problem of learning the best alternative to create a class of models known as multi-armed bandit problems. In bandit problems, rewards from alternatives are distributions rather than fixed amounts. Bernoulli Bandit Problems 伯努利老虎机问题a subclass of bandit problems in which each alternative has a fixed probability of producing a successful outcome. （未知概率）Each of a collection of alternatives {A, B, C, D,…, N} has an unknown...

Rugged-Landscape Models 崎岖景观模型（属性的集合）Like spatial and hedonic models, the rugged-landscape model defines an entity as a collection of attributes. Each set of attributes maps to a value. The goal is to modify attributes to construct an entity of highest value. （属性的相互依赖）We apply the model to reveal how interdependence in the effects of attributes makes innovation difficult and leads to path dependence in the solutions found and also leads to a greater variety of solutions. We also see how more...

Learning ModelsIndividual Learning: Reinforcement（权重）In reinforcement learning, an individual chooses actions based on the weights of those actions. Actions with a lot of weight are chosen more often than actions with little weight. （奖励）The weight assigned to an action depends on the reward (payoff) that a person has received from taking that action in the past. This reinforcement of high-reward payoffs leads to better actions being taken. （问题）The question we explore is whether reinforcement...